HOME
QUICK SEARCH
AUTHORS
KEYWORDS
HELP
BUGSSUGI 1993 Papers - Index
255 papers. Source: SAS Users Group International conference proceedings.
| # | Title× | Authors× | Abstract× | Keywords× | Extras | Pages |
|---|---|---|---|---|---|---|
| 1 | 180 Terrific Ideas for Using and Learning the SAS System | Helen D. Carey, Ginger Carey |
proc step, dataset, length statement, sample, proc freq | source | 6 | |
| 2 | A Broadly Applicable Statistical Program (BASP) for the Preparation and Analysis of Longitudinal Data Sets | Susan Y.J. Zhou, Lawrence A. Kingsley, Sheila S. H. Leung, Patricia S. Wehman |
A broadly applicable statistical program (BASP) written in SAS and SAS macros was developed for the systematic preparation and analysis of longi tudinal data sets. BASP is designed to deal with several complications of the data structure for a long-term and large-scaled longitudinal study. These complications include (1) multiple types of data files where each type has numerous, serial data files, with multiple variables; (2) the time related nature of disease development and risk factors; (3) missing data for variables; and (4) the | dataset, macro definition, weighted mean, calendar year, nominal variable, proc means | source | 6 |
| 3 | A Corporate Revenue Planning Application Using SAS/AF Software | William F. Vander Lippe, Joseph S. Bartling |
SAS/AF" software is an integral part of the development, enhancement, and maintenance of an on-line, interactive application in use at Hall mark CARDS, Inc. SAS/AF software was an easy to-leam and easy-to-use tool during the develop ment process. It is an equally useful, flexible, and adaptable tool in making enhancements as the application moves toward a production environ ment |
sas base, data cleaning | source | 6 |
| 4 | A Data-Driven Graph Algorithm Using SAS/IML Release 6.06 | Russell R. Holmes | Graphical representation of data is by far one of the hest methods that we can employ to expedite comprehension of information. Many application systems will often include a graphic subsystem to give its users the ability to plot and/or chart data in a form that not only aids understanding, hut adds to presentation value. However, when the situation occurs that involves the plotting of points very dissimilar in magnitude, problems abound. At worst, the graph for some data cannot be produced; at best, the graph is presented in a somewhat distorted display of axes and points. A graphic subsystem was developed by the CANDA group at Pfizer Inc, CSA Department for such situations. (NOTE: CANDAs are Computer Assisted New Drug Applications). Using SASIIML graphics software, a data-driven algorithm was coded which produces a relatively symmetrical X*Y plot of pre- vs. post-treatment data points regardless of their magnitude | source | 5 | |
| 5 | A First Look at SAS Software for Windows | Thomas A. Miron | This paper provides an overview of some unique features available with SAS Software for Windows. Items covered include a benchmark comparison between SAS Software for OS/2 and Windows, the display manager work environment, the SAS Toolbox window, and DOE (Dynamic Data Ex change). Benefits SAS Software for the Windows operating system provides an opportunity for Windows users to take advantage of SAS Institute products in a familiar and friendly environment. SAS Software for Windows is useful when you need to switch between Windows applications or cut and paste informa tion between a SAS display manager window and another Windows application. SAS Software for Windows also al lows you to use Dynamic Data Exchange (DOE) to link to other DOE capable applications. This feature is useful for reading data held in non-SAS files. Most options and menu functions in the AWS will be famil iar if you have used the $AS Display Manager under other operating systems | proc gmap, scatter plot | source | 5 |
| 6 | A Fuzzy Clustering Approach in Support of Identifying Influential Observations and the Computation of Multiple Outlier Diagnostics | Konstantinos Triantis, Ritu Puri, William L. Seaver |
Extensions of two existing SAS procedures are presented. The first extension OUTLIER SUBSETS is in reference to the procedure of identifying multiple outlier diagnostics for pairs, triplets and for selected subsets of an existing size. The second extension FUZZY uses the l('h nearest neighbor hard clustering procedure of Wong and Lane (1983) as input to a fuzzy clustering procedure. Both of these procedures are illustrated using the Daniel and Wood (1971) data set | outlier, regression analysis, influential observation, sensitivity analysis | source | 6 |
| 7 | A Generalized SAS Management System for Online Data Entry Applications | Roger Mueller, Priscilla E. Van Grevenhof |
The only thing that is consistent in the world is inconsistency. With new products and more computer power at our fingertips, application developers can accomplish more in less time. Furthermore, more capabilities can be given to the user in a much more standardized and generalized format. Along with the benefits for the user, the application developer gains because maintenance can become easier. At the Mayo Foundation over 60 on-line data entry systems have been developed in version 5 for medical research projects. Many of these projects were under a generalized system while others were individual entities. With version 6.08 SAS/AF, SAS/EIS, SASIFSP, and SASISCL we can now bring these systems under one generalized system with many more capabilities and options |
data science, macro variable | source | 6 |
| 8 | A GIS Is Born! | Francois Tessier, Michael Beaudry, Richard Toussaint |
This paper describes the use of SASIBASE·, SASlAP, SASIGRAPH" and SASfTOOLKIT'" In creating a research oriented geographical Information system for the City of Montreal. It discusses the production of a cartographic interface that combines the power of SASIAF and SASI GRAPH and the experience acquired by using SASfTOOL KIT in providing additlonnal .geographical manipulation procedures- such as .area description" and -area inclu sion" | proc step, data standard | source | 5 |
| 9 | A Goal Management System Using SAS/EIS | Khasha Dehnad, Matthew Casella, David Bowman |
create view, dataset | source | 4 | |
| 10 | A Macro for Multiple Stars Plot | Andy Mauromoustakos, Kevin C. Thompson |
The usual methods of statistical analysis of data from experiments in which the objective is to comJ)ilt'B treatment means for each one of se""ral response variables involve analysis of variance fallowed by some technique for means separation. Usually in most experiments the investigator measures treabnent perfonnance on several response variables. Research scientists are familiar with the process of making interpretations, and subsequently presenting the results from these methods in multiple figures, One for each response, with an error bar on the side as a guide for oompering treatments. But occaSionally the researcher may be interested in presenting treatment differences for all responses in a single graph so _ that he can gain some "a,dditional information on the relationships among the responses in addition to the treatment perfonnance. A need far graphs of this sort arises in many fields, but it is most commonly used for presenting the results in sensory evaluations studies | response variable, anova | source | 6 |
| 11 | A Macro to Make External Flat Files | H. Ian Whitlock | A SAs macro to construct external flat files from SAS data sets is considered from the view point of the user interface. The central issue is - How do you give the user as much control as he wants without requiring more work than it takes to write out the whole program? The universality of the problem and the simplicity of the ·pur· statement DATA STEP program make it a classic macro problem. The topic is presented as an example for teaching how to develop a $AS macro. A plan is laid out, the parameter interface as a system of defautts is discussed, and then the macro code is presented. With the exception of using arrays of macro variables and some optional %UNQUOTEs, only basic macro techniques are used in the code |
proc contents, proc step, dataset, flat file, numeric variable, call symput, macro loop, macro call, macro quoting function, macro conditional, macro definition | source | 6 |
| 12 | A Mainframe/Microcomputer Personnel Salary Review Input and Analysis Management Information Application | Robert A. Somer, Charles E. Pecot |
A mainframe/microcClf1l)Uter personnel salary review and analysis application was developed that incorporates the strengths of three "database" software products. Mainframe data extraction and processing is accomplished using base SAS software. Mainframe screens, menus and queries utilize the Dunn and Bradstreet Software (OBS) MilLennium System DeveLopment TooL The PC module empLoys Clipper database appLication software to provide a user friendly distributed processing environment. A variety of communications configurations are supported that enable users with virtually any connectivity to the mainframe to use the appl ication. One of the major benefits of this appl ication is the ability for users to input and analyze proposed salary data interactively on a daily basis, as opposed to'weeks of overnight paper input batch processing | batch processing, data cleaning, proc freq | source | 10 |
| 13 | A Menu System Featuring SAS/QC Software for Monitoring Alpha Detector Performance | Delano M. Splett | Statistical Process Control (SPC) can be described as a systematic method for improving the quality of production processes by analyzing, quantWying, and classifying the variation of processes with the goal of controlling and reducing the variation. It is often dWficu~ to successfully implement and benefH from SPC in a production environment due to problems such as untimely feedback of critical process changes, seemingly unmanageable amounts of data, and lack of training in SPC methodology. The implementation of user·friendly menu systems created using SAS software can be effective in eliminating these problems. This paper addresses a real·life SPC application encountered at the Rocky Flats Nuclear Weapons Plant that was solved using SAS/AF" and SAS/AC e software under the VMS operating system. Presented is a description of the application, a discussion of the development stages involved whh the project, and an |
quality control, sample, proc step | source | 3 |
| 14 | A Model for Multivariate Repeated Measures with Missing Data | Richard R. Allen | Many aulhors of statistical texts dealing with analysis of repeated measures experiments suggest the use of the multivariate analysis for data since this type of model does not require the restrictive assumptions on the variance structure that the univariate split-plot model requires. Unfortunately. the most readily available statistical programs to analyze these models require, that there be no missing data and will eliminate any observations that have missing data from the analysis. A method due to K1einbaum (1973) for analyzing the Growth Curve Model with missing data can be generalized to allow analysis of repeated measures data by a multivariate model when there is miSSing data. This paper explains this .generalization and gives a program using SAS/IML e software that can be used to anaJyze data of this type |
missing value, repeated measures | source | 5 |
| 15 | A Program Walk-through for a SAS Application | Robert G. Hall | proc step, catalog | source | 4 | |
| 16 | A SAS Macro Procedure to Summarize the Output of Logistic Regression Models | Benjamin I. Broder, Chih-Ping Chou |
A SAS program, LOGRPT. has been developed to summarize the results obtained from logistic regression. The summary presents the results from several logistic regression models. aU with the same set of covariates. but with differing dependent variables. For each covariate the following information is provided: regression weights, their standard errors, odds ratios, and a one character indication of the significance level of each regression weight. To assess the fit of the entire model, the -2 Log Likelihood statistic and its p-value are also presented. This statistic has a cm-square distribution under the null hypothesis that all of the explanatory variables (covariates) in the model are zero. The LOGRPT program also allows users to specify a subset of covariates to be reported for comparison between models. The program uses the LOGISTIC procedure as well as SAS/LML and SAS macros to summarize the results |
response variable, regression analysis, estimate, logistic distribution, logistic regression, proc calis, covariance, proc reg, predictor variable, proc logistic, significance level, likelihood function, macro definition, macro loop, macro variable | source | 6 |
| 17 | A SAS Macro to Calculate Standard Errors for Proportions from Simple Random Sampling within Strata | Wayne E. Johnson | Standard statistical programs calculate standard errors as· if the data were collected in a simple random /Sample, yielding incorrect | sample, flag variable, do over, random sampling | source | 6 |
| 18 | A SAS Macro to Perform Tests for Heterogeneity of Variances | Robert D. Abelson | Several tests of homogeneity are discussed: the Bartlett tests (chi~ square and F), Levene's test, and the Box·Andersen test. A SAS macro to perform these tests is presented | degrees of freedom, anova, sample | source | 3 |
| 19 | A SAS Solution: Building a Flexible/Standard System to Report the Analysis of Efficacy Data for Clinical Trials | Judy A. Sullivan, Naitee Ting |
clinical trial, macro variable, macro definition, lookup, proc format, macro loop, proc step, %if-%then-%else | source | 6 | |
| 20 | A Simulation-Based Project For Experimental Design Based on SAS Software | Michael Friendly | This paper describes a simulation-based course project have used for some years in an intermediate-level undergraduate course in regression and experimental design. The proj.eel -is designed to simulate the statistical and the practical issues in real research in' as many ways as possible 9 within the constraints of conducting the project in a large class. The student's task is to design, execute, and analyze one or more studies of a substantively real experiment. "Data collection" consists of specifying the number of groups, factor levels. and number of observations per group. The student runs a (hidden) SAS program, which suppnes the student with data simulated from a model known only to the instructor. Each student is given a "research budget", each experiment has associated costs, and students are rewarded for each correct statistical decision | sample, sample size, power analysis, main effect, data science, outlier, dataset | source | 8 |
| 21 | A Six Sigma Control Chart | Meng-Koon Chua | A Six Sigma control chart is proposed as a cross functional management tool and illustrated with an example. The benefits of using this control chart and the role of SAS programming are described. The fundamentals of Defects Per Million Opportunities (DPMO) and Defects Per Million Units (DPMU) are introduced, and the relationships between these units for attribute and variable data are discussed. In addition, the | sas programming, standard deviation | source | 5 |
| 22 | A Tutorial on Logistic Regression | Ying C. So | Many procedures in SAS/STAT~ can be used to perform 10' gistic regression analysis: CATMOD, GENMOD,LOGISTIC, al)d PROBIT. Each procedure has special features that make. ~ useful for gertain applications. For most applica· tions, PROC LOGISTIC is the preferred choice,. It fits binary response or proportional odds models, provides various model·selection methods to identify important prognostic variables from a large number of candidate variables, and computes regression diagnostic statistics. This tutorial dis· cusses some of the problems users encountered when they used the LOGISTIC procedure |
logistic distribution, likelihood function, regression analysis, logistic regression, proc step, sample, proc logistic, maximum likelihood estimation, predictor variable | source | 6 |
| 23 | Adding Convenience to the SAS System: One Site's Implementation Strategy | Valerie Hilton | When Tandy Information Services became a SAS site, the SAS System was new to most of the 200 plus members of the programming staff and the user community had very little mainframe experience. The system needed to be implemented in such a way that it was convenient to use and it minimized the learning curve associated with the mainframe environment. The system also needed to be standardized to lessen the burden on the support staff | proc step, macro definition, production run, sample | source | 6 |
| 24 | Administering a Large, Long-Term Clinical Trial Requiring a Hierarchical Database Using the SAS System, Release 6.07 | Mitchell Snyder, Mark Goldberg, Michal Benderly |
The Israeli SOCiety for the Prevention of Heart Attacks is conducting a clinical trial involving over 37,000 patients at 18 medical centers during the screening period, of whom 3,000 are selected for a five to seven year follow-up period. Some of the 8 different types of forms may appear only once per patient, while others may appear as many as 42 times. We also store | randomization, adverse events, data science, sample, clinical trial, quality assurance, dataset, screening period | source | 6 |
| 25 | Aesthetics/Artistry | Susan J. Kenny | source | 2 | ||
| 26 | Alternatives Within the SAS System | Sherry Joyce King | source | 7 | ||
| 27 | An Application of SAS Software in Auditing and Finance: Monetary Unit Sampling | Pankaj Patel, Laura G. Bacon |
This paper describes an interactive use of SAS for the auditing and finance area. Monetary Unit Sampling is described in detail to show SAS's efficiency in programming and sampling. SAS's built-in functions and procedures make the program efficient and easy to understand. The program was written and developed on the IBM-PC, MSDOS, using base SAS for the personal computer version 6.04 and IBM 3090, MVS, running on base SAS Version 6.0 | sample, binomial distribution, sas base, sample size, dataset | source | 6 |
| 28 | An Automated System Using SAS Software for Tracking Radiological Environmental Monitoring Data for the San Onofre Nuclear Generating Station | Ellen A. Curry | So.uthern Califo.rnia Ediso.n's Co.rpo.rate Policy and federal regulatiens both require radio.lo.gical enviro.nmental mo.nito.ring fo.r the San Onofre Nuclear Generating Statien (SONGS). Results of the Radio.lo.gical Enviro.nmental Mo.nito.ring Program (REMP) co.nducted at SONGS are then published annually. A SAS system was designed which reperts ambient radiatio.n levels measured in a 50-mile radius of SONGS. Items sampled fro.m the environment included air,crops, water, seil, sho.reline sediment, no.n-migrato.ry marine species and both po.table and ecean water. The resulting radio.lo.gical monito.ring data utilized a co.mputer system to. sto.re and repo.rt the data generated by the samples taken. This paper reperts how SASI AF applicatio.ns are used to. create the Menus (with Extended Table Pro.cessing), Data Entry Screens, Data Validatio.n, Repo.rting, Graphics, and Utilities. Individual SAS Datasets were created fo.r each of the databases to. respo.nd to. their unique requirements | sample, proc print, dataset, proc step | source | 8 |
| 29 | An Automation Procedure to Create Codebook and SAS Files from Questionnaires | Eric Yu I. Wang, Chih-Ping Chou, Meliessa M. Jordan |
Generating codebooks from questionnaires and developing an SAS program to create SAS files with variable labels and format after data has been collected is usually a very time consuming and error-prone procedure. Ideally, these procedures can be accomplished by usjng the information directly from questionnaires. A program GENCF, has been developed in this paper to integrate this process without manually typing codebook and SAS source codes to create SAS file. This automation procedure may generate more accurate products more efficiently | character variable, label statement | source | 4 |
| 30 | An Environmental Database Using Character Data Storage | Nat H. Wooding Jr. | Some environmental measurements that would usually be consi dered to be numeric are often found to be a number and some modifying symbol. Measurements below a detection limit are an example. Alternative methods of storing these data are discussed and a method for reducing some storage require ments is offered. Virginia Power is an electric utility with 12,500 megawatts of generating capacity located in Virginia, West Virginia, and North Carolina. A consequence of the generating process as well as other company oper ations is environmental monitoring required by various federal and state statutes. Data collected by the company's Water Quality.Department come from soil, streams, wells, and power station discharges. Parameters recorded include field measurements such as pH, temperature, and dissolved oxygen levels, species counts and other biological measure ments, and results of quali tative and quantitative labor atory analyses | numeric variable, sample | source | 5 |
| 31 | An Exact Confidence Interval for the Ratio of Means Using Nonlinear Regression | Tapon Roy | In several statistical applications, the ratio of means is of primary interest. The construction of confidence intervals for such ratios can be done using various techniques, most of which are less than satisfactory. In this paper, a nonlinear regression model will be used to calculate an exact confidence interval for the ratio of means. This result will be compared to a commonly used approximate teChnique, and to the empirical~istnbution based bootstrap method. A pharmacokinetic example involving the bioequivalence of formulations will illustrate the methodology, and a SAS"" program to perform the calculations will be presented | confidence interval, bootstrap, regression analysis, proc nlin, sample, correlation, proc step | source | 6 |
| 32 | An Extraordinary Reporting System for the Pharmaceutical Industry - the Macro Way | Judy L. Gilbert | Automated Clinical Data Output, ACDO, is an Eli Ully and Company project that has been dedicated to developing ftexible, sophisticated SAS macros which are used to create output across all disease classes. SAS macros are developed in a modular design so that multiple reports may use many of the same modules. The primary focus of the reports has been demographic, adverSe event, and laboratory data, but some .of the modul~s also analyze or summarize drug efficacy data. Although the listing programs will handle any clinical trial design, including crossovers, the summaries are generally applicable only fOf parallel designs. A fundamental and powerful a\tribute of this reporting system is flexibility. The SAS .macro facility is fully utilized. All reports contain multiple· nested macros, SAS procedures, and functions. Customer-specified parameters number from 37 in the simplest report to over 200 in a more complex report | clinical trial, adverse events, proc step, proc glm, macro variable, pharmaceutical industry, proc genmod, data cleaning, anova, proc rank | source | 5 |
| 33 | An Integrated Clinical Trials System Utilizing Client Server and Graphical User Interface Technology | Martin J. Rosenberg | Motivated by the desire to better manage an ever increasing volume of information and to reduce the length of time required to introduce new drugs around the world, the pharmaceutical industry and regulatory agencies such as the U.S. Food and Drug Administration have sought to facilitate the drug development and approval processes through innovative uses of computer technology. One such effort that has received great attention is the computer assisted NDA or CANDA and its cousin for biologicals the computer assisted PLA or | clinical trial, crf, data review, clinical data, clinical research, pharmaceutical industry, key, data cleaning | source | 7 |
| 34 | An Interactive SAS/AF System for Sample Size Calculation | Deborah J. Matos | In the design of a clinical trial, the number of patients selected is typically . determined through use of statistical criteria. Sample sizes are cho.sen to ensure that a clinical effect of some pre specified magnitude can be detected with sufficiently high probability. Calculation of sample size with provision for control of Type I and II error rates is an essential part of the planning of an adequate and well-controlled clinical trial. This paper describes PCSIZE, a SAS/AF system, which makes the calculation of sample size estimates fast and easy. This PC-based application was developed under PC-SAS' Version 6.04 (for MS-DOS') using Base SAS', SAS/AF, and Screen Control language" (SCl). First, some basic background theory on sample size estimation is presented, and then focus is shifted to the interactive SASI AF system and its features |
sample size, sample, null hypothesis, parallel group, significance level, clinical trial, type i error, degrees of freedom, response variable | source | 6 |
| 35 | An Introduction to Compressing Data Sets | Juliana M. Ma | character variable, length statement, options statement | source | 5 | |
| 36 | An Introduction to PROC GREPLAY | Marc Jacobs | proc step, proc greplay, proc gmap | source | 7 | |
| 37 | An Introduction to SAS/FSP Software | Cyndie B. Gareleck, Terry J. Fain |
SASIFSP is a set of procedures used to perform full-screen interactive data entry, data editing, data retrieval, and data query. This tutorial provides a basic overview of SASIFSP software under version 6.0716.08 of the SAS System. Topics include: creating a new SAS data set customizing data screens data entry data query browsing SAS data sets and external files The focus of this tutorial is on PROC FSEDIT although other procedures such es PROC FSBROWSE and PROC FSVIEW will be briefly discussed. Screen control language (SCL) is considered an advanced topic and will not be covered in this tutorial |
proc step, dataset | source | 4 |
| 38 | An Introduction to SAS/LAB Software | Katherine J. Roggenkamp | ancova, covariance, confidence interval, data visualization, anova, main effect | source | 10 | |
| 39 | Analysis of Split-Plot Experiments Using SAS Software | Stephen P. Jones, Monica Y. Chi |
The SASISTAT procedures, PROC ANOVA and PROC GLM, will conduct an appropriate analysis of split-plot designs if the experiment is replicated or if there are higher-order interactions that can be used as valid estimates of the error terms. Frequently, split-plot designs are run without replication and the error terms are derived by pooling together higher order interactions; a pooling that cannot be accom plished readily using current SAS procedures. In this paper we will describe SAS programs that have been developed to conduct an appropriate analysis of unreplicated split-plot designs |
residual, anova, proc step, randomization | source | 6 |
| 40 | Analysis-of-Variance Computing: Then and Now, with Reference to UnbalancedData | Shayle R. Searle | A brief account is given of calculating analyses of variance, progressing from mechanical multiplying machines to successive digiting on tabulators to today's high~ software. Implication of the latter for the analysis of unbalanced data is discussed, together with a summary of that analysis | anova, main effect, proc genmod, interaction effect | source | 12 |
| 41 | Analyzing Multiple Response Survey Data: An Overview of Strategies and Methods | Paul E. Pope | This paper addresses the difficulties associated with analyzing multiple response survey data with SAS' software. Two existing strategies of analyzing multiple response data are examined: the multiple dichotomy strategy and the multiple response strategy. After reviewing the existing method for implementing the multiple response strategy (contained in the SAS' Applications Guide), three alternative multiple response | proc freq, proc step, frequency analysis | source | 6 |
| 42 | Application of Conditional Logistic Regression Methods to Event History and Multiple-Choice Data | Dan Steinberg | sample, likelihood function, response variable, regression analysis, logistic regression, logistic distribution, panel data, residual, maximum likelihood estimation, proc step, conditional logistic regression, proc phreg | source | 4 | |
| 43 | Application of SAS Software in Analyzing Strategic Network System Output: A Case Study at a Leading Brokerage Firm | Ashoke Bose | In this paper, first I shall discuss a conceptual model of an Infonnation Flow Simulator & Management System (lFSMS) in light of communications management problems as encountered in today' s small and large cOIporations. FolloWing this, r shall provide a brief description of the Strategic Network (SnetPlus) System software which is an innovative and powerful IFSMS developed by Communication Software, Inc. (CSI). In order to demonstrate the practical use of the IFSMS, I shall present a SnetPlus model which was developed 10 simulate the flow of infonnation in the client-prospect network of a fmancial consultant at a leading brokerage fum. Further, I shall present the results of the analysis of the output of the SnetPlus model by SAS software and evaluate the probability of success of selling the broker's services | sample, proc step | source | 6 |
| 44 | Applications of the IML Procedure for Multiple Response Permutation Tests | William D. Johnson, Donald E. Mercante |
This article discusses statistical methods for the analysis of multivariate data in experiments with a relatively small number of subjects. Methods based on the assumption of multivariate normality have power that decreases rapidly as the number of variates increases for a fixed sample size. In some investigations, the number of variates exceeds the sample size creating an awkward dilemma in classical multivariate analysis. Permutation tests have many useful applications especiaJly where the assumptions such as normal distributions and hornescedasticity are in question. Additionally, the number of variates relative to the sample size is not an issue with these tests .. We discuss a multivariate permutation test and present a readily implemented IML procedure for constructing these tests. Software for the multiple response permutation tests was available previously in the SAS Supplemental Library as PROC MRPP but has been discontinued in SAS Version 6 |
permutation test, sample, sample size, multivariate analysis, euclidean distance, proc step, proc iml | source | 5 |
| 45 | Applying Quality Control to Retail Scanner Data | Joseph S. Bartling | proc autoreg, proc step | source | 2 | |
| 46 | AUTODOC - An Automated SAS Programming Documentation System | Kenneth P. Wilson | A problem that programmers in many industries face is deciding on the most expedient method to document programs, internally and externally. This poster demonstrates a SASe Macro that facilitates simple but thorough internal documenation of programs, as well as SAS· program documentation database maintenance for external documenation. The macro was designed to read and store program header information, reducing programmer time necessary to create an audit trail for a specific project. Historically, information used to be entered into a similar system manually, but can now be automatically input from the program header. Also demonstrated are a variety of reports that can be produced utiHzing the database that this macro updates and maintains. The reports include simple layouts for project archiving, as well as more complex reports such as program flow diagrams necessary to recreate an entire project | audit trail, i/o | source | 4 |
| 47 | Automatic Features for Row, Column, and Missing Value Handling for a Macro That Builds Tables from FREQ Procedure Output Files | Hugh Geary | proc step, proc freq, macro call, do loop, keyword parameter, %macro parameters, macro loop, macro definition | source | 6 | |
| 48 | Bayesian Statistics in SAS Software | William J. Raynor Jr. | Recent advances in computational statistics have greatly simplffied the implementation of Bayesian analyses in the day-to-day practice of statistics. This paper provides an overview of Bayesian methods, comparing and contrasting them with classical meth ODS, and emphasizing the central role of the likeli hood function in parametric analyses. We will outline some of the recent computational methods, particu larly the Gibbs algorithm, Sampling-Importance Re sampling, and Data Augmentation, and show how they can be used to generate posterior distributions for many quantities of interest. Examples will include normal theory models (including constraints) and bi nary regression techniques. The algorithms to be discussed make extenSive use of the SASTM macro language, SASlIMLTM, and SAS DATA steps to aug ment the resutts of SAS PROC steps |
sample, likelihood function, posterior probability, regression analysis, macro put, monte carlo, logistic distribution, markov chain, proc step, prior probability, proc freq | source | 10 |
| 49 | Benchmarking PROC REPORT and PROC TABULATE for Complex Tables | Felix G. Meale, Robert G. Hall |
Differenc.es in coding techniques were examined between PROe REPORT 'and PROe TABULATE. Six programmers with varying degrees of experience were asked to code three different tables of varying complexity. Results between programmers arld procedures were compared. By far the most significant difference in execution times were due to poor data management programming rather than between procedures. In general, however, TABULATE ran faster than PROe REPORT | proc step, proc report, proc tabulate, cpu time | source | 7 |
| 50 | Beyond Column Input: Advanced Techniques with the INFILE and INPUT Statements | Kathy McLeod, Nancy Mae Bonney |
The < routine SASs programming techniques for reading raw data files are the list, Column and Formatted modes of the INPUT statement. While the standard use of these modes will usually satisfy many data requirements, the exceptions require that the SAS programmer understand how to go beyond the "default" use of the INFILE and INPUT statements. In addition, version 6.07 of the SAS system offers additional INFILE statement options and INPUT statement features that many SAS programmers have not had the opportunity to discover. In this paper we will discuss the limitations of the "default" approaches, and present examples of advanced programming techniques using the INFILE and INPUT statements that can be used' to overcome these limitations. Included in this tutorial will be explanations of new features such as the use of the DELIMITER option for List input mode, the use of the Tilde and DSD options to read delimited data, and an explanation of the often overlooked -"Named" input mode | input statement, list input, raw data, formatted input, infile statement, named input, column input, character variable, length statement, missing value, sas log, variable, dataset, date variable | source | 10 |
| 51 | Biometrics Software: What Tools Have Pharmaceutical Biometricians Got in Their Kits? | David J. Shannon | clinical trial, sample size, sample, pharmaceutical industry | source | 6 | |
| 52 | BLOOD PRESSURE AND HEART RATE vs TIME | Lan Tran-La | proc step, proc gplot, vital signs | source | 8 | |
| 53 | Building an Executive Information System (EIS) without SAS/EIS Software | Stephen Davidson, Gregory A. Smith |
dataset, batch job | source | 5 | |
| 54 | Building and Maintaining End-User Applications | Jamie Robbins, Khasha Dehnad |
Applications that are the most important to any organization are thOse-that are used heavily. The- application use can be by. a few very influential people, or by the masses on a regular basis, The applications can help the users either track. enter, or investigate important -information. The applications need to be robust but also address the specific areas intended:' Developers of these applications need tools available to them to make them more productive. They need tools for fast prototyping as well as tools that give them totai control over every aspect of production applications. These tools need to be ones that the applications developer will not outgrow as the sophistication of the applications grow. Thl.s paper highlights a case study of the Depository Trust Company, a national clearinghouse for the securities i_ndustry, which developed an EIS to automate and control a management by c;>bjective (MBO) program | data cleaning, dataset | source | 3 |
| 55 | Building Data-Driven Object-Oriented Applications: Selected Techniques | Michael Gilman, Donald J. Henderson, Martha F. Johnson, Merry G. Rabb, David S. Septoff, Norman Shusterman, Gregory A. Smith, Robina G. Thornton, Philip H. Van Dusen, C. Jessica Yuan |
ThIs tutorial descr1bes a systam's arch~ecture In which an appHca· tion is buitt using reusable moc;Iuies or ciesses of objects. The appfi· cation can be buitt. controlled. maintained. and enhanced through changes to the under1ying data and parameter files. An application designed In this way allows for more f1exibiUty and lower mai",,.. nance than one where the code is written to perform specific func· tions. Generalized moc;IUles or classes of objects. the actiOns of which are dictated by parameters. lend themselves to the inevitable changes that are made to the system. Many moc;Iifications can be accomplished by changing values In parameter fiies instaed of changing the source code. Features of this approach include adding new reports changing or reordering menu Items * controlUng user access and security. A sample application that makes use of SASiAP and SASlFSP" software using Screen Control Language (SCL). the SaL pr0ce dure | sample, proc step, macro call, macro variable, proc sql | source | 8 |
| 56 | Building the World's Largest Local SAS Users Group: An Organizationist Manifesto | LeRoy Bessler | source | 3 | ||
| 57 | Cars, Dogs, and Drivers: Working with Large "Public "Data Sets | Joseph E. St. Sauver | sample, dataset, cpu time, raw data, proc format, proc step | source | ||
| 58 | Centralized SAS System Management: Just Do It! | Suzanne S. Scott | Managing the SAS System at a large Corporation, such as Texas Instruments (TI), offers many challenges. Different sHes, spread across the nation; different SAS System product configurations, from bare minimum to full-up; and different platforms, from mainframe to PC, are just a few of the many challenges to be faced. At TI, several procedures have been established to ease this large administrative task, provide comprehensive support, and, of course, save. money. Detailed spreadsheets are used to keep customer information and product configurations up to-date; contract summary sheets, called 'Release Statements', help maintain legal compliance; SAS System NEWS bulletins keep customers current on new developments; and licensing consolidations across the corporation allow major cost savings. Policies such as these have made it possible for a small staff of three to manage and provide for the needs of the 300+ SAS System customers at TI | sas base, access control | source | 5 |
| 59 | Chart Smart: Designing Graphs and Tables for Effective Communication | LeRoy Bessler | bar chart, horizontal bar chart, line chart | source | 9 | |
| 60 | Clinical Data Processing Efficiency Techniques | Denis Michel | Efficiency may be defined as conservl;!iion of resources in performing I;! II;!Sk. Optimizing .the use of computer resources is important in a clinical data processing environment because large data sets are routinely manipulated and then analyzed and reported in batch mode. Program maintenance time also needs 10 be minimized because SAS code is often copied, modified, and shared by programmers, statisticians, and other users. This paper presents programming techniq~ for improving performance on most platforms without attempting to quantify efficiency through benchmarking. The techniques are illustrated using specific clinical I'flSeIlI'Ch applications | length statement, proc step, missing value, numeric variable, cpu time, dataset, character variable, data step, proc contents, clinical data, libname statement, laboratory data, proc sql | source | 6 |
| 61 | Comprehensive Usage of PROC FSEDIT Version 6 | Andrew P. Ford | This paper presents an advanced application of FSEDIT under version 6. This paper shows how to: Link customized help screens, built with PROC BUILD, to an FSEDIT screen for both overall general help information as well as for individual entry field help. Create customized commands. Link a selection menu built with PROC PMENU with a FSEDIT screen. Test for entry errors and display appropriate error messages. Link together two FSEDIT screens to provide a tool for previewing output before printing a hard copy. Link customized PF key settings for both FSEDIT and OUTPUT windows |
proc step, sas output, macro definition, macro variable | source | 6 |
| 62 | Computing and Plotting First-Order Spatial Autocorrelation Statistics Using SAS | Larry J. Layne | Presence of spatial autocorrelation in classical statistical models violates the assumption of independence and leads to biased estimates of variation for tests used in inferential statistics (Griffith 1988). The SAS program presented here computes 2 indices of spatial autocorrelation: Moran's I and Geary's c ratio, and their respective tests of significance. In addition, the program computes the spatial mean, spatial median and standard distance, and plots these values on the areal distribution of attribute values. The program assumes a spatial arrangement of a regular lattice of squares, in any configuration, in order to compute the connectivity matrix. The program requires 3 variables in the data set: two variables corresponding to the X-Y location and a variable containing the attribute, or response, measurement, for each areal unit | correlation, numeric functions, sample | source | 6 |
| 63 | Cooperative Processing with the SAS System | John Mclntyre | One of the four fundamental "barriers to successful enterprise~wide information delivery is the incompatibility of various hardware and operating system environments. Customers withenterprise-wide information needs must be able to leverage their investment in technology, both old and new. Many terms are used in the field to describe the strategies available to customers including client/server, distributed processing, and downsizing. This paper will examine_models of coo~rative processing, review the use of SAS System tools-in particular SAS/CONNEC~ and SAS/ACCESS software-to accomplish the integration of resources, and suggest some guidelines for the distribution of application tasks |
relational database, stored procedure | source | 4 |
| 64 | Creating Flexible Reporting Applications Using Si\SjAFSoftware | Candy R. Habich | sample, where statement, proc sql, create view, proc step | source | 5 | |
| 65 | Current and Future Developments for SAS Software under MV5 | Brian Bowman | This paper discusses MVS specific developments in Release 6.08 of the SAS System and current research activities. The primary focus is on connectivity, interoperability, and the integration of SAS Software with other environments | sas base, decimal | source | 3 |
| 66 | Current and Future Developments for the SAS System under CMS | Mary Fenno | In this session, members of the eMS host development group will discuss current research and development projects. Recent enhancements to the eMS environment offer new and exciting possibilities for the SAS System under eMS. Discussions will include high-speed servers, multi-tasking, Shared File System exploitation, and connectivity | sample, sas output | source | 5 |
| 67 | Data Compression and the SAS System | Philip Bardsley | It seems inevitable that no matter how much disk space we start with for our work. we always need more. With disk space at a premium, we need to take advantage of ways to store more data in less room. This paper investigates the pros and cons of space-saving tools that are available on four different computing platforms: an IB~ mainframe running the MVS· operating system, an IBM RiSe Sys tem/SOOO· workstat-ion running the AIX- operating system, a microcomputer running the MS-DOS operating system, and a microcomputer running the OS/2- operating system. The SAS software tools investigated are the COMPRESS= option, which is available in Release 6.07 of the SAS Sys tem running on MVS and AIX and in Release 6.08 on OS/2. and the SAS LENGTH statement. which is available on all platforms. The tools outside the SAS System. here called system compression tools, are the AIX compress' command and the shareware programs PKZIP f'or DOS machines and ZIP for OS/2 | length statement, decimal, cpu time, libname statement, raw data, missing value, dataset | source | 6 |
| 68 | Data Entry Using Extended Tables in Release 6.07 of the SAS System | Allan M. Kominek | This paper discusses a system for entering the results of a survey. The information for this particular survey relates to the background of people in the Cleveland SAS Users Group. Through the use of SAS/AF, Screen Control Language (SCL), and extended tables, a system was developed that displayed the question and all possible choices, including blank lines for comments. The data entry person could then select the answers from the list and enter any comments. The entire system required one screen and three SAS data sets |
source | 6 | |
| 69 | DATA Step Changes and Enhancements | Jeffrey A. Polzin | dataset, do loop, modify statement, output, by group processing, proc step, select, set statement, master data management, proc sort, leave, update statement, automatic variable, composite key, sample, proc print | source | 6 | |
| 70 | DATA Step Kappa Computations | Peter P. Gaccione | This workshop leads users through the data manipulations necessary fo calculate the weighted kappa statistic (Spitzer, Cohen, Fleiss, and Endicott, 1967). As a generally accepted measure of rating reliability, kappa can be used to estimate an amount of agreement among raters. In this presentation, data from two physicians who rated films produced by magnetic resonance imaging .(MRI) will be analyzed. In addition a SAS program utilizing DATA STEP statements will be shown, and its dissection will indicate the flexibility SAS allows in programming within the DATA STEP for calculations not readily available from procedure output. Emphasis will be placed on the program commands and how the data is set up to perform the computations. Output from the PRINT procedure will be used extensively |
proc print, proc step, numeric functions, proc freq | source | 6 |
| 71 | Date Manipulation Techniques, or How D'ya Show a Date a Good Time? | Jerry Kagan | This paper. is about date manip!liation techniques. The main thrust will be a systematic approach to handlin~ partial dates within the existing framework of SAS date values. A history of the calendar system is given as an introduction to provide background. The concept of partial dates as usable infonnation is examined. Sample macros for reading and printing partial dates are provided in the appendixes. The ability to use partial dates is of particular interest in the pbannaceutical industry where incomplete dates are often encountered, but the concepts presented here would apply to any industry with a need to work with incomplete date infonnation | date variable, sysdate9, macro conditional, macro loop, macro variable, macro definition, macro put, missing value | source | 10 |
| 72 | DBCS and NLS Support in the SAS System | Randy Whitehead | This paper describes new and ongoing efforts to provide oecs anc::t NLS support within the SAse System. Areas covered include expanded translate table support allowing users to provide tables for upper and lower casing 0:1 data, local-to-transport and transport· to-lOCal processing. interpretation of local national characters as SAS special characters (u-umlaut for right-curly-brace, etc.), and others | proc step, array, proc report, options statement | source | 8 |
| 73 | Death Matching - An Algorithm and Comparison of Its Implementation in PROC SQL versus PROC SQL plus a DATA Step | Charlotte Corelle, Dean MacLaughlin, Lois Drew, Mary Longacre |
This paper describes how we used base SAS software to match person-based records to other person-based records without having the benefit of a unique identifier on which to merge. It illustrates how analysts at the Center for Health Research (CHR) in Portland, Oregon, linked records of a study population whose vital status was unknown with state death certificate records to determine mortality. CHR analysts modified a death-matching algorithm used at the Boston Collaborative Drug Surveil lance Program and implemented it in two different schemes, both of which used PROC SQL and a DATA STEP. One of the coding schemes was considerably more efficient than the other. This paper presents the matching algorithm and compares the perfonnance of the two coding implementations |
proc sql, proc step, where statement, unique key | source | 3 |
| 74 | DELPAGE: A SAS Macro to Delete Page Breaks in SAS Output | Victor Kamensky | SAS outputs from different procedures go to different pages. Sometimes an output ,fi le has -too many smat l pages. Pages in the output fi le -ar,e separated by page break symboLs in the first column. Macro DELPAGE deletes all unnecessary p$ge break symbols | sas output, character variable, dataset, keyword parameter, %macro parameters | source | 3 |
| 75 | Demystifying Extended Tables | Stephen Davidson | The SASlAF" product allows SAse users to Cl88te powerful inter active tables called extended tablas for infonnation display or enby. However, cl88ting an extended table can, at times, seam to be a very imposing and mysterious task. This tutorial will deal with the when, why, and how 0' extended tables. We'll see when H is appropriate to use one, why use one, and how to build one. We'll bwld bOth. introductory and advanced tables along the _ as examples. This papar assumes a besic knoWledge 0' the SASiAF Scraen Control Language, and will be use'ul 'or people who have never used extended tables before or those would like to under stand them better | create table, create view | source | 8 |
| 76 | Designing What IS Really Asked For: Volleyball Tournaments and Nonorthogonally Blocked, Split-Plot Fractional Factorials | Diccon R. E. Bancroft, William D. Kahn |
In this tutorial we show how to use SAS to design experiments which combine both blocking structure and split-plots. The standard theory of incomplete block designs assumes no structure on its treatments (PROC PLAN). Standard fractional factorials come with very restrictive block sizes (PROC FACTEX). We meld these traditions in a rich and realistic manner. Most industrial experiments are split-plots. Material from one (main-plot) experiment becomes the blocks for another (split-plot) experiment. We show how to simultaneously fractionate both the main-plot factors and the split-plot factors. Consider how you might design the following real experiment: Study the effect of four two-level factors and their six two-factor interactions. Normally we would do this with a 24, but in this situation the material comes in blocks of size 3. We have enough material for 13 such blocks. These 13 blocks themselves consist of a 23 design with 5 replicated points |
proc step, degrees of freedom, proc sort, stage 2 | source | 7 |
| 77 | Developing an Interactive Method of Displaying Pavement Management Information | Bruce Dietrich | Highway engineers must analyze large volumes of roadway data as part of the process of maintaining our highway system. An interactive Pavement Man agement Reporting System has been de veloped at the Florida Department of Transportation using SAS/AF and PROC GMAP to facilitate this process. Tech niques were 'developed to dynamically access and process the data and to display the information as color coded geographic roadway maps. The system uses GMAP features such as unit areas, response data sets, annotations, leg ends and titles to generate the cus tomized maps based on user selections. The application uses numerous SASjAF program entries to allow user .selected branching and to facilitate modular program development and modification. Pushbutton fields arid pop up selection lists are used to create graphical user interfaces so the system can easily be used by non-programmers |
proc step, feature selection, macro variable, proc gmap | source | 8 |
| 78 | Developing Mainframe Applications in a PC Environment | Steven J. First | Quite often developers don't have mainframes onsite, but do have PC systems with mainframe connections. SASe software offers an unusually high degree of machine platform independence and as such offers an excellent opportunity to develop applications on personal computers and later port those applications to mainframes. This paper will share experiences from several successful major implementations and wil address issues such as types of connections, uploading and downloading of programs, reports, SAS catalogs, and graphics, both with and without SAS/CONNECT. software. Differences and pitfalls when moving between platforms will be highlighted as well as presenting several utilities developed for intercepting mainframe spool output, translating mainframe carriage control, test file generation, reading of EBCDIC flat files in an ASCII environment and vice versa, automatic generation and testing of input statements, and more |
input statement, dataset, proc step, catalog, sample, proc download, sas output, proc print, macro variable | source | 9 |
| 79 | Developing User-Friendly Applications Using the sAs System for PCs | Caroline C. Bahler | Application development in the SAS system for PCs can be accomplished by using e~her the MACRO language in BASE SAS or SAS/AP' software. The developer's choice of· the SASI AF software or the MAC RO language to produce the application depends upon several factors: a) the objective of the applica tion; b) the amount of PC pardware "personaliz<:ttion" needed; and c) the time frame allowed to produce the application. Applications developed by uSing the MACRO language are easily "personalized" for each PC hardware configuration and can be very complex or very simple. However, the popm'enus and function keys olthe SAS system for PCs are not availableto the _ program. developer. Applications. developed .~slng SASI AF software have the advantage of greater user friendliness" through the use of Help screens and function keys but need to be hardware independent |
dataset, macro loop, sample, macro variable, macro conditional, macro facility, macro definition | source | 6 |
| 80 | Development of an Analysis System Based on the Information System of a Social Organization | Habibollah Bin Haran, Mohd Yunus B. Majid, Faridah Maaral, Noriah B. Idris |
The development of information system has been given much attention lately. Realising this phenomena, the need to develop an analysis system seems appropriate - as this system would efficiently assists decision makers in solving certain pertinent issues. This paper discusses the development of such system based on pre existing information system. SAS procedures like SAS Base, FSP, AF, STAT, ETS and GRAPH are used. The case study chosen for the development of this system is one of the largest and most active social organisation in Malaysia and the software used by this organisation to keep important information is dBaseIV. Thus the development of this system includes two phases namely the linking of dBaseIV and SAS and the development of an analysis system. Keywords: Information System, Analysis System, System Development, procedure, SAS and dBaseIV | proc freq, histogram | source | 7 |
| 81 | Discovering the FUN in SAS Functions | Neil H. Howard | put statement, proc step, input statement, missing value, lookup, character functions, numeric functions | source | 5 | |
| 82 | Distribution of Scientific Data via a Wide Area Network: A SAS/AF User Interface Based on VMS | Debra A. Battista, Jeffrey S. Rosen, Richard W. Latimer |
The Environmental Monnoring and Assessment Program (EMAP) was developed by the Environmental Protection Agency (EPA) in cooperation wnh a number of other federal agencies. The objectives of the program are to estimate the current status, extent, chang!!s, and trends in indicators of the coridnion of the nation's ecological resources. Large amounts of data are collected each sampling year which are verified, validated, and organized into SASe datasets. The estuaries group of EMAP has designed and developed a prototype data distribution system which provides access of data over the EPA Wide Area Network (!NAN). The prototype system was· developed in response to requests for EMAP estuaries data by several major user groups. Addnionally, the development of this system tested the access of data stored on different nodes of a WAN | dataset, cpu time, proc sql | source | 6 |
| 83 | Double-Key Data Entry/Verification in SAS Software: Part of Downsizing a Clinical Data System to the OS/2 Operating System | Barry R. Cohen | The pharmaceutical industry, which has typically used SAS software only for the data analysis phase of clinical data processing, now has a new opportunity to use SAS software for the data capture phase. The history of the use of SAS software by this ind\lstry is briefly reviewed, explaining why this new opportunity exists today. Next, a powerful double-key entry/verification system, which was built in SAS software to capitalize on this opportunity, is described. Finally, issues of porting this mainframe system to the OS/2 platform are addressed | clinical data, data wrangling, filename statement, sas base, data science, pharmaceutical industry, proc step, data cleaning | source | 7 |
| 84 | Downsizing a Mainframe SAS Application: Issue of Porting to the OS/2 Operating System | Barry R. Cohen | The SAS System's multi-vendor architecture is a valuable asset for application developers. Developers can build a SAS application on one platform, and then port it to another without concern for whether or not the application can run in the new environment. The SAS Institute has already assured us that it will! This is heady stuff, indeed, for those who have been around a while and have struggled through earlier-era system migrations where absolutely everything was a stumbling block. With all the deserved hoopla about SAS's comprehensive compatibility across platforms, though, one might assume that a downsize project is now relegated to one of building SAS transport files for the application code, moving them across platforms, converting them back to SAS system files, and then changing a few LlBNAME and FILENAME statements. But this view is misleading in many instances. Often, this is only the beginning of the journey. But do not be daunted |
sas library, filename statement, clinical data, proc step | source | 6 |
| 85 | DSGEN: Magically Generate SAS Data Sets, Customized Screens and Formats Without Writing Source Code | Bernadette H. Johnson, Bonnie Duncan |
sas base, label statement, variable, data cleaning | source | 5 | |
| 86 | Effective Data Graphics with SAs/GRAPH Software | L. M. Harschnitz, Laurie A. Abell |
proc step, bar chart, horizontal bar chart, histogram | source | 6 | |
| 87 | Effective Use of the MULTTEST Procedure in Japanese Biopharmaceutical Studies for Drug Development | Masakazu Andoh, Tohru Uwoi |
In March 1992, the guideline for statistical analysis of clinical data has been notified in Japan. This guideline suggests several points of multiple statistical testing issues. Many types of multiplicities have been arising from clinical trials. Especially one of the characteristic issues in Japan is recategorization of ordinal responses ill contingency table. MULTTEST Procedure is a new procedure in release 6.07 SAS/STA~ software. It offers flexible approaches to the multiple statistical testing issues, p-value adjustments by the resampling methods, bootstrap and permutation. I! is computationally intensive but very attractive(Hamada and Yoshida, 1992). The purpose of this paper is to evaluate the availability of MULTTEST Procedure to deal with some issues of multiplicities in Japanese biopharmaceutical studies for drug development |
proc freq, bootstrap, chi-square test, clinical trial, permutation | source | 5 |
| 88 | Emergent Visual Information Systems | Anthony R. MacKay, Stuart A. Nisbet |
dataset, pharmaceutical industry | source | 2 | |
| 89 | END;END;END;END;END and Other Useful Shortcuts Using Function Keys | Louise Weiler | Shortcuts are the. stuff of dreams for computer users. The SAS System provides many opportunities for shortcuts, one of them being function keys. One key stroke can execute a SAS command or a set of com mands .. One keystroke can also make a text-string magically appear on your screen. Function keys can thus be big time savers for you and for users of applica tions thai you develop. This paper will serve as a guide to understandi~g, defining, and managing function keys. Youwililearn the source of default function key settings, and you will learn many ways to change these defaults. Finally, this paper will present some useful function key settings collected from SAS users. This paper was developed in the MVS environment, but will be relevant to all SAS users | proc step, proc print | source | 6 |
| 90 | Enhanced Support Services from SAS Institute's Technical Support Division | Sally J. Painter | At SUGI 17, the Technical Support Division announced three new support services available to SAS software installations. These three new features, in addition to the existing Online CUstomer Support Facility, offer a full range of electronic support services. The electronic support services available are: Electronic Mail Interface to Technical Support (EMITS) Online Customer Support Facility (OCSF) TSNEWS-L File and Mail Ust Anonymous FTP These services allow SAS SoftoNare Representatives, Consultants, and users to communicate with the Technical Support Division electronically. This paper will discuss the services with an emphasis on the best way to utilize each. There will also be a discussion on other enhancements to Technical Support services | sas log, sample | source | 3 |
| 91 | Enhancing Data Entry Screens for the Processing of "The Directory of Texas Manufacturers "Using SAS Screen Control Language, SAS/AF, and SAS/FSP | Betty Smith | The processing of the Ilirootory of Texas Manufacturers (DTM), wllich has been produced by the B""",u of Business Resealch since 1933, bas always been lengthy and somewhat complicated. Over the past few years, funding cuts have forced reductions in staff whicb even more complicate the process. Luckily, the advent of Screen Control Language witb the release version 6.07 for CMS bas been a great help in alleviating some of the pressure in producing the 1993 edition. Using Screen Control Language (SCL), we have been able to do key word searcbes of reference data sets wI!icb help with Standard Industrial Classification (SIC) coding, perform automatic abbreviations of words in fields such as the fmn and personnel title fields which conform with DIRECTORY standards, allow the entry operator on-line access to the DTM data sets for reference in processing entries, among other tasks which seem to automate the entry process | raw data, where statement | source | 6 |
| 92 | Enterprise-wide Report Writing | Jeff Babcock | In this paper I hope to identify the issues surrounding delivering an enterprise-wide reporting strategy. Upon identification of the issues I hope to provide a framework for a solution to the issues by utilizing the SAS System for Infannation Delivery | dataset, flat file | source | 3 |
| 93 | Estimating Neural Network Nonparametric Production Functions Using SAS Software | Daric L. Brummet, Lawrence C. Marsh |
neural network, network optimization, null hypothesis, likelihood function, sample, proc step | source | 5 | |
| 94 | Estimation of the Lorenz Curve and Concentration Ratio | Shi-Tao Yeh | macro function, proc nlin, macro definition | source | 5 | |
| 95 | Experiences with a General SAS Macro for the Construction of Classification and Regression Trees | Roland Brandmaier, H. Ansari, D. Dickson |
Clinical studies are rarely designed wHh the main aim of determination of prognostic factors. Such analysis is usually exploratory in nature and may be helpful in hypothesis generation and design for further trials. For this analysis common classical regression methods are used. An attemative method is the identification of pro gnostic factors by classification and regression trees (CART, Breimann et al. 1984). While regres sion models may be used to assess the impact of theoretically independent risk factors on the outco me variable, the classnication tree technique is concerned wHh prognostically homogenous sub groups of patients. The latter method works on the principle of separating the population into groups in each of a series of steps. /Id each step the popu- '1ation splH using each potential prognostic variable is 'examined and the "best" splH is selected. The terminal nodes of the tree indicate the prognostic subgroups which can then be ranked | regression analysis, proc step, clinical trial, censoring, sample size, macro definition, sample | source | 4 |
| 96 | Exploiting Personal Computer Operating Systems with the SAS System | Carol Williams Rigsbee | This paper describes enhancements to the SAS System in Release 6.08 that allow the exploijation of the features available under OS/2 and Microsoft Windows operating systems. The goal of the paper is to show users of the SAS System under OS/2 or Mi crosoft Windows the features and customi zation options that enable the SAS System to work natively on PC platforms | excel, proc step | source | 7 |
| 97 | Exploratory Data Analysis Using SAS Software | Greg J. Weier, Randall D. Tobias |
This tutorial describes two environments within the SAS System for carrying out exploratory data analysis. SAS/LAB software provides a guided, interactive environment for users who lack an extensive statistical background. Analysis results are presented graphically with interpretations, warnings about assumptions that maybe violated, and suggestions for further analyses. SAS/LAB software focuses on basic methods commonly requested by researchers and engineers. SAS/INSIGH-Y@ software provides a highly interactive graphical environment in which windows are a means 'of viewing and manipulating data objects. Scatter plot smoothers and numerous parametric curves with associated diagnostic plots can be displayed. Tools, palettes, and pull-down menus enable the user to further explore objects. Windows are dynamically linked so that actions on observations in one window are reflected in all windows |
correlation, scatter plot, proc genmod, simple linear regression, histogram, regression analysis, proc princomp, covariance, proc reg, ancova, box plot, data science, outlier, binomial distribution, probability distribution, response variable | source | 11 |
| 98 | Facilitating Clinical Data Management and Regulatory Review in the Pharmaceutical Industry: A Manager's Perspective | Randi Olsen McFarland | DespiIe advances il computer technology, !he management of clinicallrials data has remained a challenging task. The compelilNe naIur8 of our indIslry, combined wi!h!he quan~ty and variety of infonnalion coIlecIBd, creales a need to enter data in a way !hat faciitates subsequent retrieval, review, and analysis. EnhallClOlTlSnlS IoWId in SAS Version 6.06 and higher provide !he requiremenls for a cIaIabase and review syslam suitable to !he pharmaceutical indus1ry's needs. With !hese oonside!ams in mind, Genelabs decided to Iiosnse !he ClinAccess'" clinicallrials system, a !hird-party system buift wiIh SAS/AF software. ClinAccess lakes acWanlage of !he recent enhancements in SAS software to simplify !he data enlry, review and analysis precesses and !hereby enhances our progress towanls regulatory submission. ClinAccess also provides !he ability to review data which laciitates in-house review by !he clinical staff |
data cleaning, regulatory submission, adverse events | source | 2 |
| 99 | Facilitation of Retrieval of Brain Image Measurements with the SAS System | Gail Lerner, Nancy Y. Li, Martin Redmond, Houwei Wu, Jose Maria J. Alvir, Robert Bilder, Manzar Ashtari |
We are in the process of devel oping a database management system for users of a Magnetic Resonance Neuroimaging Labora tory. The information from the contiguous slice brain images are output directly into an ASCII file, in a UNIX environ ment. This occurs as soon as the clinician finishes making determinations about the brain subsections. The large amount of data produced by this system has created a challenge for a team of researchers, biostatis ticians, and a computer scien tist as we set out to facili tate the retrieval of this in formation with the help of the SAS System. The importance of maintaining information about the contiguous slice images is relayed by their role in struc ture definition enhancement, as well as 3D images | lag function, substr function | source | 4 |
| 100 | Fastest Route Problem Using Weighted Shortest Route Technique | Mohd Yunus Maji, Faridah Maara, Syed Othmawi A. Rahman |
The sh~t path problem in a tranSpon netw~rk. usually concerns with determining the sh0!lcst distance between a SOIltCC and a des~ation. The problem is normally faced by police dcparuncnt. fIre department and hospItal emergency services where immediate ~ervices arc fe9.uired from them at any given time of the day. The problem concerned Involves ~ultl.ple sources, fony arc:~ and, tyo'CDty four destinations. The algorithm developed m this model uses SAS OR m deciding which route to be used at different time of me day. A real case study was done in the city of Kuala Lumpur, Malaysia to assist the fire departJ?ellt. police and hospital in deciding which route to follow at a given time of the day. A wClgh.ted sho~t ~ut~ problem was .designed to ensure that the time required to reach the deSU'ed desttnatlon IS the shortest ume (and not necessarily the shortest route). Constraint like traffic intensity is included in the model | sas output, proc network | source | 3 |
| 101 | Fighting Over the Pie: A Base SAS Application to Forecast Market Shares | Jon Rosenthal | J.D. Power & Associates, a market r~search firm, produces 'a forecast of car and truck sales each month. There are about 300 different vehicles. Each is forecasted by month, for five_ years beyond the current year. However each vehicle must be individually treated as to such things as capacity. promotion, introduction, phase-out, seasonality, etc. This is in addition to economic factors and historical trends. The challenge was to create a robust, automated system that handles a tremendous amount of data and computations, yet easily allows dynamic, user input and control during its execution, is self-contained, ,and produces extensive and standardized output. This system uses only base SAS. It runs on a 486 PC under OS/2. This paper demonstrates several techniques of dealing with system limitations from a large job, such as unavailable memory and storage | macro variable, normalization, sas base | source | 6 |
| 102 | Filling in the Blanks: Multiple Imputation for Replacing Missing Values in Survey Data | P. M. Wright | Multiple imputation Is a useful technique to apply to survey data with missing values because It allows the use of complete data analysis methods whlch mlghtnotbe applled otberwlse and It allows one to Incorporate variance between imputattons as well as wlthm the dataset. This paper presents a review of common imputatton methods. shows how touse SASe to Compute multlpleimputaUons using regression methods. and outlines howto compute con1ldence Intervals wbich Incorporate both wltbin and between imputation variances | imputation, missing value, multiple imputation, proc step, regression analysis | source | 3 |
| 103 | Forecasting Financial Performance for Long-Term Care Facilities with SAS Software | William S. Calvert | A strategy is presented for forecasting fmancia! perfonnance of Long Term Care facilities. A simulation program is described and segments of the financial forecasting program are shown as implemented in SAS. The input, output, and flow of:the sample program is 'described. Some coding technique is highlighted, including changes necessitated by moving to the PC environment. The problems and features of more complex forecasting situations are also discussed | sample, proc step | source | 6 |
| 104 | Forecasting Methods | Mark Little | This paper is a tutorial on time series forecasting methods. It pro vides a brief survey of time series forecasting tools in the SAS Sys tem, principally in SAS/ETS software |
proc esm, time series analysis, moving average, proc step, regression analysis, sample, proc forecast, confidence interval, binary variable, goodness of fit, ma, autoregressive model, forecast horizon, feature selection, proc gplot, seasonality, proc arima, proc reg, model validation, seasonal difference | source | 13 |
| 105 | Forecasting the Volatility of Financial Markets: ARCH/GARCH Models and the AUTOREG Procedure | Minbo Kim | standard deviation, sample, proc step, proc autoreg, interest rate, time series analysis, residual, regression analysis | source | ||
| 106 | Frequency Tables Using the SAS System: What They Are and What They Could Be | Grant Blank | Frequency tables seem simple. yet the SAS System is unable to produce acceptable tables; problems include: severe Omits on the length of variable labels and fonnats. no ability to store extended 18xt describing· variables (like the text of quesHons). and HmHed ability to handle multiple response variables, among others. As a result many intensive usets of frequencies, Oks survey research finns, use their own custofTWIritten software. This paper describes aspects of a project underlaken for the National Opinion Research Center (NORC) to produce the NORC Automated Codebook. " uses the concept of a codebook to consider how the characteristics of data and data analysis are changng in the 1990s. Greater ability to describe and document datasets and variables is becoming important for scientific, marketing, business, and govemment data | proc freq, missing value, label statement, proc step, dataset | source | 4 |
| 107 | Generating Confidence in the SAS System: Confidence Intervals for Commonly Used Procedures | Jose Maria J. Alvir | Confidence intervals have recently been galnmg favor as the preferred manner of reporting statistical results in medical journals. Correspondingly. there has been a trend away from overreliance on probability (P) values. A P value informs the reader about the probability that the population parameter being estimated using sample data is nil. combining information on both the magnitude and the precision of the parameter estimate. The P value depends on sample size. Two large groups may be significantly different from each other in spite of the observed difference being small and clinically insignificant. On the other hand, wo small groups may not be significantly different from each other in spite of a large observed difference. Confidence intervals separate the information about the magnitude and the precision of the parameter being estimated. Reliance on P values may in part be due to the emphasis placed on P values in statistical sofl\.vare packages | confidence interval, sample, correlation, p-value, estimate, sample size, tolerance interval, hypothesis testing, proc step | source | 4 |
| 108 | Generic Applications Development Using the SAS:Macro Language | Christian M. Forster | The capabil ities and imovative techniques of two generic, real-world, SAse Macro Language reporting applications used internationally by a major pharmaceutical finn are presented. Together, these applications can handle the majority of listings and tables found in a typical cl inical subnission. Users of these appt icat10ns find their report turnarOU"lCl time reduced to a fraction of the time traditional methods used to take. The results they now achieve are more flexible, consistent, reproducible, aesthetically pLeasing, and more accurate than results users accCJq)l ished using conventional SAS prograrnning, or worse, hand-typing. Generic reporting systems are errphasized, but the techniques presented have a wide variety of uses. Exanple caLLs to GENLIST, which is a generic listing system, and GENTABLE, a generic table-producing system ·(written by this paper's author and Arrr( Frueh, also fram DZS) are presented along with some macro uti 1 hies. Using our | macro variable, variable, macro definition, proc step, macro facility, sample, proc freq, put statement | source | 6 |
| 109 | Getting to Know You, Getting to Know All About You: Simple Techniques for Forming a Relationship with Your Data | Arthur L. Carpenter | Statistical techniques such as cQrrelation analysis, regression analysis, analysis of variance, and others are used to establish relationships among data variables. When some effort is taken, these relationships can often be seen QY exploring the data using less sophisticated techniques prior to performing the analysis. These data exploration techniques often have side benefits by exposing aberrant data values and hidden or unanticipated relationships. This workshop will examine 80me simple techniques for getting acquainted with your data. This workshop will use the DATA STEP and simple summary PROCedures 'to detect bad or incorrect data points, find data outliers~ notice patterns, determine frequency distributions, and establish relationships among data val;ables. In general the workshop participant will be introduced to techniques that can be used to get to know the data. We become acquainted 'with our data in much the same way that we become acquainted with another person |
proc step, raw data, unique key, regression analysis, proc format, proc print, proc reg, proc univariate, standard deviation, numeric variable | source | 6 |
| 110 | Getting What You Want Out of the TABULATE Procedure with Pre- and Post-Processing | Aaron L. Dukes | proc step, proc tabulate, class variable | source | 6 | |
| 111 | Graphical Evaluation of Product Characteristics | Melissa A. Durfee | Utilizing SAS/GRAPH, many aspects of product characteristics may be displayed and analyzed. Besid.es using regression plots to examine relation ships between variables, contour and 3-D response surface plots may be constructed on two independent variables. In designed experiments, potentially significant factors may be indicated by Bayes posterior plot of effect estimates coupled with a cube plot to examine the response. In a Taguchi experiment, the optimal design may be identified through a main effects plot using the appropriate signal-to-noise ratio. In addition, the nature of product variation may be assessed using multi-vari (multivariate) ,&Iots. By interfacing SAS/GRAPH with SAS/QC procedures, process stability and capability may be evaluated. These graphical applications facilitate and simplify the use of statistics in industry |
regression analysis, contour plot, proc step, main effect, response variable, histogram | source | 6 |
| 112 | Graphical Presentation of Multivariate Longitudinal Data | Maria L. Shabe | The analyses of Clinical trials often involve the examination of complex longitudinal data. Frequently, in such analyses, it is useful to attribute the temporal behavior of one variable with that of another. Consequently, the presentation of synchronized data becomes critical, as is the need to simultaneously display data from different treatment groups. As an ex"mple, in a randomized clinical trial designed to assess the effectiveness of four treatment regimens in patients with seasonal rhinitis, treatment efficacy was likely to be confounded by measurable but uncontrollable environmental factors such as pollen counts, precipitation, and humidity. 959 PROC GPLOT in SAS/GRAPH can be used to display data of such complexity. By using the. PLOT2 statement, both the efficacy measure (daily mean severity score) and a confounding factor (pollen count) can be displayed simultaneously. Step 1: Prepare the Data Set Delete variables with blank values |
proc step, clinical trial | source | 2 |
| 113 | Guided Data Analysis with SAS/LAB Software | Katherine J. Roggenkamp, Wayne E. Watson, Randall D. Tobias |
SAS/LAB@ software is a data analysis advisor for en gineers and scientists who lack an_extensive statistical |
data visualization, dataset, outlier, response variable, scatter plot, data science, anova, homoscedasticity, main effect, regression analysis | source | 7 |
| 114 | Hindntarsh Electoral Division: Analysis of persons aged 65 and over | Andrew Kopras | proc step, bar chart, proc greplay | source | 11 | |
| 115 | How to Implement a Graphical User Interface Using the SAS System, Release 6.06 under MVS (without Confusing Your Present Users) | James D. Hefner | IBMs 8AA strategy provided a common graphical interface for all programs on all platfonns. Known as CUA, this is the interface that is implemented by SAS software under MVSfTSO, OS/2 and other platforms, starting with version 6.06. However, the full CUA implementation under MVSfTSO assumes that the user has a graphics terminal with a mouse. Many of the mainframe terminals in use today do not have a mouse, or even graphics capability. In acoommodating users who have such terminals, a programmer using SAS/AF has to design each SAS Application so that the interface does not confuse their non-mouse users, while still being useful for mouse users. This paper outlines the steps in creating a sample application. The main menu screen and help screens are discussed and built so that they are context-sensitive. A program dalog box is constructed, showing how to design the entry fields so that they guide both mouse and non-mouse users in the entry of data |
sample, proc step | source | 10 |
| 116 | Identifying Sales Regions Using PROC GMAP | Deb Cassidy | proc step, proc gmap, proc freq | source | 5 | |
| 117 | If Nested IF-THEN/ELSE Statements Give You Headaches, Then Try ARRAY 3-D-or Else Take Asprin | Judy A. D. Alicante, Glenn D. lronhat |
Have you ever spent more time indenting and aligning nested IF-THEN/ELSE code than writing it? After writing two pages of the same DO-END block or starting your lines past column forty, do you wonder if there isn't a better way? There is a better way with SAS version 6.07. You can replace your Stone Age code with a New Age three dimensional array. ARRAY 3-D shows how we re placed 75% of IF-THEN/ELSE logic with a decision matrix. The decision matrix was then converted into SAS code as a three-dimensional array. The three dimensional array construct increases program read ability and maintainability, in addition to relieving your IF-THEN/ELSE headaches | if-then-else, transaction data | source | 6 |
| 118 | Instant Makeovers: Changing the Appearance and Functionality of the X Interface to the SAS System | Marianne O'Connor | catalog, sas base | source | 6 | |
| 119 | Integrating Decision Support | Jo Fisher, Marco Mongalo |
One 01 the hallmarKs 01 the SAS' System is i1s wide varletyol analyt· icaI tools: descriptive statistics. mu~ivar1ete techniques. lorecasting and other time serieS analyses. metheinatical analysis, statistical quality ImprOvement. as well .. project management end linear pro gramming. This peper provides a case study 01 how Royal Carib bean Cruises Ltd. has integrated a number 01 analytical techniques ilto a oomplete decision support system - their Yield Management srs | demand forecasting, quality control | source | 5 |
| 120 | Internetworking for the Beginner | Janice Kolb, David F. Kolb |
SAS/CONNECT software is a cooperative processing tool which allows a local SAS session to utilize files, hardware resources, and SAS software on various remote-hardware platforms such as mainframes, mini~computers, workstations and personal computers. You can establish a SAS/CONNECT conversation to multiple SAS sessions on multiple hardware platforms to distribute SAS processing to the most appropriate machine in order to use your computing resources to their best advantage. To be able to use SAS/CONNECT software, all of the-different hardware platforms have to be networked together. In'ternetworking can be defined as "the communication between data processing devices on one network and other possibly dissimilar devices on another network" |
error message, traffic data | source | 7 |
| 121 | Introduction to DATA Step Views | Jeffrey A. Polzin | In early versions of SAS·software, SAS programmers used DATA STEP programs to access data. stored in external files. If the data required analytical prooassing from SAS procedures, users oreated a SAS data file. The DATA STEP view provides an altemate method of accessing and managing data transparently. DATA STEP views inherit the power and flexibiity of the DATA STEP with the ability to manipulate and manage data from a variety of sources including external files and existing SAS data files. This paper describes the input and output DATA STEP conceptual models. application uses of each type. as well as defines the creation process for each view type. A small application is used to demonstrate how one may easHy convert an existing application to use DATA STEP views. Perfor mance issues of CPU. 110, elapsed time, and memory utilization are discussed together with comparison data using eertier data extrac tion methods |
dataset, proc step, modify statement, proc sort | source | 5 |
| 122 | Is It Possible to Print More Than One Graph on a Page? A Comparison of PROC GREPLAY and the DATA Step Graphics Interface (DSGI) | Lynn M. Thome-Polingo | This paper will show two methods for p'rinting more- then one graphic output on a page. Alsd, the paper will show the ability to surpass the realm of FROe GREPLAY and to end with the unexpected ability to create customized diagrams with the DATA STEP Graphics Interface (DSGI), to meet the standards of critical viewers. Still, GREPLAy'is suited for quick and easy use through its windowing features. The objective of this 'paper is to explain and illustrate, the differences between these two tools. This way one knows the advantages 'and limitations to consider when selecting one versus the other as a tool for replaying graphic. IN'rRODOCTION: There exist two methods in SAS which allow the user to play back graphs stored in graphics catalogs. These methods are PROC GREPLAY and the DATA STEP Graphics Interface (DSGI) tool~ Both methods are very simila-r, but DSGI has a greater capability to customize- output. This paper will begin to discuss how the two |
proc greplay, proc step | source | 6 |
| 123 | Is the SAS System a Database Management System? | William D. Clifford | Commercial Database Management Systems (DBMSs) provide applications with fast access to large quantities of data. In addition, many have other capabilities such as data integrity services, data sharing, application-creation tools, and report writing. Version 6 of the SAS System also contains a number of similar features. This paper examines the database features of the Version 6 SAS System and compares them to the services offered by several popular DBMSs. The | constraint, flat file, data cleaning, where statement, audit trail, data standard, proc sql | source | 6 |
| 124 | Is The System Analysis Phase Really Necessary? | Faith Reneé Sloan | Most systems deveLopMent projects are I.I1CIertaken with the expectation that 20 percent of the ti .. will be invested in the analysis phase and 80 percent in the implementation phase. Whlt.ny organi;zations are conthKJOusly faced 'With is the increase in resources needed to maintain and enhance these systems. Spending a lot of time in the analysis phase is IIOst tillleS said to be -:infeasible or illpf"8Ctical. The plain truth is that many projects fail due to the unwillingness to spend more time initially defining systelll requirements. IMTllCllUCTIa. System analysi,$ is the process of extracting the needs of a system~"'what the systelll lUSt do to satisfy the user, not how the system will be inpLeMnted. Figure 1 illustrates a macro yiftj of the system deveLopIent life cycleC,SDLC) approach to the syste-. rec:fJi reMents and 'evaluation or the systera analysis phase | proc step, data dictionary | source | 6 |
| 125 | Learning SAS by Diagrams and Examples | Chao-Ying Joanne Peng, Shu-Yeng Wong |
This paper outlines an alternative method, of teaching SAS to computer~,inexperienced lind sUitistic:s-phobic students in social sciences. SpecJfically we are interested in demonstrating how the $AS system and j,ts programming capability can be easily understood through diagrams and examples. We believe diagrams are so effective visually that they assist students in conceptualizing the structure and logic inherent in $AS programming language. Hence. they' help to reduce logical errors often mllde by beginning--Ievel users who lack training in any formal programming langu8ge. wmples of short $AS programs, on the other hand, can be used as templates by students. These templates teach students to ·spe'ak-the SAS language correctly. thus, minimizing the likelihood of making syntax errors. This alternative method of teaching has been tried in an introductory SAS course at Indiana University for four years. Studen1S whc completed such a course readily demonstrated an ease | dataset, proc step, data science, error message, conditional do loop | source | 6 |
| 126 | Likelihood Ratio Tests for Regression Coefficients from the Cox Proportional Hazards Model Using the PHREG Procedure | Laurent M. Kassalow | The PHREG procedure provides a TEST statement for testing hypotheses for regression coefficients from the Cox proportional hazards model. PHREG presents Wald tests for the test statistics. Many authors prefer likelihood ratio test statistics for testing these hypotheses. A SAS macro will 'be presented that performs likelihood ratio tests for these hypothesis tests. This macro can also be used to perform stratified analyses |
likelihood function, regression analysis, proc phreg, predictor variable, macro call, macro variable, maximum likelihood estimation | source | 6 |
| 127 | Linking AS/400 Databases with PC-DOS SAS Software | Caroline C. Bahler, Katie Pentz, Sue Colyer |
Using data collected in AS/400 databases by PC/SAS requires a two step approach. First, the information must be downloaded from the ASl400 system and then analyzed by programs written with SAS software. An application was developed at Supelco, Inc. to query and extract information from an ASl400database, download the extracted information to a SHARED FOLDER, analyze the data, and then append data to two SAS datasets stored on the same SHARED FOLDER. This application is run daiiy with a report generated on a daily basis and control charts generated once a month. AS/400 database query, extraction, and downloading is accomplished through an IBMI!I PCSUPPORTIfl program which configures a PC as an ASl400 workstation and part of the AS/400 drive system as a DOS environment through SHARED FOLDERS. The advantage to using a SHARED FOLDER is that the data is password-protected through the AS/400 system so sensitive data Is not readily accessible and hard drive space is freed up | dataset, infile statement | source | 3 |
| 128 | Linking Databases with Base SAS Software to Facilitate Client/End-User Access to an Extended Data Resource | Philip J. d'Almada, Cynthia J. Berg |
needed by matching records from two databases. The master database"w8s a large SAse data set stored on mainframe disc. The supplemental database. was a larger raw database stored by year on two tapes. To meet this need, the authors developed a sophisticated, iterative linking program in base SAS software. The program, using three to five linking processes, was applied twice because of annualized data, and each application was dichotomized conditional on a key linking variable being either knownlnonrnissing or unknowrv'missing. Record linking required the SAS data set option. IN::, in a MERGE statement, and three or four linking variables in various combinations in a BY._. statement, and was conditioned with IF . THEN OUTPUT and ELSE *** -statement-pairs. Linked records were passed through a validation filter for an overall 99.7 per cent acceptance rate. Linkage validation used four SAS functions. INDEX, SCAN. SUBSTR. and TRIM in IF | data cleaning, dataset, sas base | source | 4 |
| 129 | Macros Can Condense and Simplify Statistical Output for Presentation to End Users | John A. Wenston | The power and wealth of statistical procedures available in SAS can satisfy almost any analytic computing need, but the volume of information provided can be overwhelming, particularly for presentation to end users or consumers. Users often want only a few summary measures, e.g., odds ratios or adjusted means, from the volumes of output SAS·generates. Tailoring information to consumer needs often involves wading through the output and constructing summary tables manually. The use of macros can save a substantial amount of the time and effort that this process often requires. This paper will present guidelines on how to use the SAS macro facilny to create parameter driven programs that take the results of statistical procedures,calculate ·summary measures no.t provided by the procedure, and present the output In tabular form | proc step, proc print, dataset, logistic distribution, regression analysis | source | 6 |
| 130 | Mailing Large PC SAS Files via BITNET | Samuel D. Calhoun | The use of large PC SAS· files has been increasing at the Economic Research Service. Until recently t.he SAS trade data had to be broken down into smaller files for analysis. Due to changes in the trade agreements the need to do more bilateral trade research increased. This demand created larger PC SAS files. This made it more difficult to transport large SAS files from PC to PC or Mini to Mini. The BITNET" communication environment has eased the large PC SAS data transportability problems on our VM/MVS Mini computer. Since BITNET is a form of electronic mailer, it can communicate directly to other users on the system and other MVS systems such a Universities and other government Agencies | macro definition, macro loop | source | 3 |
| 131 | Management of Time-Range Data | Andrew A. Norton | proc step, create table, proc sql | source | 8 | |
| 132 | Migration and Growth: Generating Small Area Population Projections Using SAS Based Multiregional Modeling Techniques | Louis Schwarz | An organization's computerized customer data base system has generally lent itself to the routine processing of transactions such as new entries, withdrawals, and address changes. All of these routine and mundane transactions are recorded, but seldom viewed as part of a large system of customer level changes (geographic and/or status). If one views these transactions as being analogous to the traditional demographic equation [Population (Time 2) = Population (Time 1) + Births (Time 1-2) - Deaths (Time 1-2) + Net Migrants (Time 1-2)], then "births" become new entries, "deaths" become withdrawals, and "migrants" are simply address changers who have moved to another geographic market. By taking a demographic approach, applying population techniques to a customer database population, one can take advantage of one of the more advanced (but seldom used) technique in demography, the multiregional population growth matrix | markov chain, dataset | source | 6 |
| 133 | Model Comparison in the One-WaY,Between Subjects Design: A SAS Implementation | John F. Walsh | A statistical test is developed based on the comparison' of sum of squared errors associated with two Competing models. A model based on cell means is compared to a representation that spec~ies the means for the treatment conditions. Comparing models is more general than the traditional H, in ANOVA that assumes all the cell means are equal. The test statistic, Proportional Increase in Error, is computed usinglhe SAS statistical system. Tests of hypotheses involving several independent means are typically analyzed as a one-way between-subjects ANOVA. The null hypothesiS, H" evaluated states that all group means are equal. The test statistic for this hypothesis is the F ratio of Ms"tMS w' Recent presentations of design (Judd & McClelland, 1989, Maxwell & Delaney., 1990) have sought to broaden the formulation of H, by using a model comparison approach. In such as | anova, proc step | source | 2 |
| 134 | Model Estimation with Inequality Constraints: A Bayesian Approach Using SAS/IMLSoftware | Charles B. Hallahan | Economic models derived from economic . theory are generally acompanied with various sets of equality and inequality restrictions. For example, in an economic relationship between consumption,c, and income, y , c. = a + Pv. + et' the coefficient p represents the marginal propensity to consume and should lie in the interval [0,1). Treating the parameters as fixed and using constrained least squares, estimates can be obtained, but the complex distributional properties of the estimates complicate inference. The Bayesian | monte carlo, likelihood function, sample, posterior probability, proc model, regression analysis, proc step | source | 7 |
| 135 | Modular Programming for a Screen Control Language Application | Linda M. Quinn, Allan M. Kominek, Tina M. Sluga |
This paper discusses the development of a system for loading, checking, and correcting the data from a clinical study. The data was cOllected using a set of five Case Report Forms (CRF). The objective of the project was to develop a system that would allow display and editing of data on screens that resembled the pages of the CRF. Error processing was similar, if not identical, for many of the CRFs. Common modules could be employed for validation routines among screens and programs. Although the system consists of over 20 screens, the actual time of system development was greatly shortened through the use of common modules | macro variable, data cleaning | source | 6 |
| 136 | MONOSEL: A SAS Macro for Model Selection in Linear Regression Analysis | Victor Kamensky, Ilya Muchnik |
A new criterion is suggested to select model in Linear Regression Analysis. The criterion is based on Quasi convex Set Functions ( Monotone Systems) theory. The algorithm is very simpLe and provides the global minimum of the criterion. C~rison of our criterion with REG ,procedure criteria on a linear combination of two variables show that our criterion_tends to select the only variable which contri,butes more into the dependent yari,able, while all REG procedure criteria keep both variabLes. Simulation studies show that for artificial data - linear function with smalt adjustment and random noise - our,macro MDNOSEL ( up to certain value of noise) always gives bigger percent of correct solutions than STEPWISE method. For real data from different sources results of our macro are compatible with REG procedure results. INTROOUCTlON Model selection problem in regression analysis is widely described. Reviews can be found in ( Hocking, 1976; Thompson,1978) |
proc reg, proc step, regression analysis, sample size, sample, feature selection, response variable, predictor variable | source | 6 |
| 137 | MONTE-A Risk Analysis Program | Richard H. Fleming | This paper describes procedures which can be wed to establish a relationship between the service that a computer center provides, and the expenses associated with providing a particular level of service. These procedures are embedded in a Monte Carlo simulation model (MONTE). The MONTE program can be used by investigators to investigate the financial effect of service level changes | batch job, predictor variable, standard deviation, response variable, monte carlo, sample, data analyst | source | 6 |
| 138 | Moving Your Names Without Losing An Alias | Lary Jones | One of the terrifying nightmares of switching systems is losing all the mail addresses one has lovingly ac quired over the years. At Binghamton University we are anticipating that many of our CMS users will be migrating to UNIX. In CMS one stores e-mail nick names and addresses in a NAMES file. The elm mail program in UNIX uses an aliases file to build a database of addresses. Of course, the format of these two files is substantially different. This paper describes a SAS program which builds an aliases.text file from a given CMS NAMES file. While an effective utility could have been written in , Pascal Or <, SAS was chosen since it is the application language I use most frequently, and the project pro vides a good example of the power of the DATA STEP language for reading delimited fields and string manip ulation |
conditional do loop, character variable | source | 5 |
| 139 | New Features of PROC NETFLOW: Using IBM's Optimization Subroutine Library (OSL) | Trevor D. Kearney | This paper describes new features in PROe NETFLOW. a proce dure in SAS/ORC!! software. These features can decrease solution times of your optimization problems. PROe NETFLOW can now use Optimization Subroutine Library (OSL) to solve linear and network programming problems. For network programming problems, you can enlist OSL to do some or all optimization. The input data, output of PRINT statements and solution data. sets are the same whether you use PROC NETFLOW's optimizer or OSlo OSL is much easier to use when controlled from within a SAS procedure. This paper will illustrate how PROe NETFLOW can interactively run OSlo |
proc step, network optimization | source | 8 |
| 140 | Object-Oriented Graphical User Interface (OOGUI) Using FRAME Entries in SAS/AFSoftware-Part I | Bill Powers, Jeff Cartier |
The FRAME entry is a new catalog entry available in SAS/AF" software. This entry enables you to develop graphical user interface applications using object-oriented programming (COP) techniques. In this first of four tutorials, you willieam about the set of supplied GUI building blocks (widgets). how to place these widgets In a FRAME entry. and how to construct a simple menu-driven application. Additionally, you willieam how to write simple Screen Control Language (Sel) programs to control execution of the application |
catalog entry, dataset, proc step | source | 5 |
| 141 | Object-Oriented Graphical User Interface (OOGUI) Using FRAME Entries in SAS/AFSoftware-Part II | Jeff Cartier | The SAS/AF FRAME entry offers many widgets that can be creatively combined to implement nearly any GUI design. New SCL features aHow widgets to communicate with one another in novel ways. This paper illustrates variations in FRAME design and new Screen Control Language features that control widget interaction |
catalog entry, dataset | source | 4 |
| 142 | Object-Oriented Graphical User Interface (OOGUI) Using FRAME Entries in SAS/AFSoftware-Part III | Jeff Cartier | The SAS/AF" FRAME entry employs Object Oriented Programming (OOP) techniques to create GUls. This paper introduces several COP concepts and how they have been implemented in the FRAME entry. Examples are provided to demonstrate how application developers can create new software units (classes) provide default settings for class attributes define new class attributes (instance variables) write generic Sel programs (methods) for classes |
catalog entry, dataset, data cleaning | source | 6 |
| 143 | Object-Oriented Graphical User Interface (OOGUI) Using FRAME Entries in SAS/AFSoftware-Part IV | David J. Biesack | This tutorial continues the discussion 01 object oriented programming when developing SASJA~ FRAME entries and other object oriented applications with SeL. Examples will show the following features of the software: Create custom subclasses of SAS/AF FRAME widget classes that provide custom behavior and reduce programming; Write custom attribute screens so that you can modify the BUILD procedure and reduce application programming; Send information from one object to another; Create utility classes and objects |
catalog entry, if-then-else, data cleaning | source | 8 |
| 144 | Once Upon a Time There Was a Menu | Marge Scerbo, John P. Elrod |
dataset, proc step | source | 7 | |
| 145 | OS/2 SAS Server: Using SAS Software in a Networked Cliet-Server System | Allan E. Williams, David A. West |
For users of the SAS System 6.04 on PC networks, the release of the SAS System 6.06 for the IBM OS/2 operating system brought both exciting opportunities and accompanying problems. The SAS System 6.08 for OS/2 is much faster, as well as allowing larger jobs and new capabilities. However, upgrading all SAS users to OS/2 is impractical for many environments. This paper discusses one solution that enables all SAS users on a PC network to take advantages of the new features and throughput of the SAS System 6.08 for OS/2 without having to convert operating systems. A simple approach to implementing an OS/2 SAS server on a network is described. Examples of the use of the SAS server are described and future enhancements and possibilities are discussed. Extensions from a single OS/2 SAS server to multiple servers on the network are straight forward. Multiple batch servers create a 'cluster' of processors that appear to the user like one large server | proc step, %if-%then-%else, macro definition | source | 6 |
| 146 | PC Data Conversion: Avoiding the Pitfalls | Daniel R. Bretheim | dataset, proc step, date variable, numeric variable, sysdate9, missing value, character variable, format statement | source | 6 | |
| 147 | Performance and Capacity Planning in a Mainframe Environment | Thomas G. Confrey, Ellie Rosenbaum |
SAS is a very efficient tool to analyze and report on performance and capacity elements in mainframe environments. This paper will concentrate on what data is readily available and some effective ways to present it to various levels of management. The techniques and rationales discussed here are not limited to mainframe environments only, but can be used as guidelines for reporting on other computers and components as well | regression analysis, capacity planning | source | 6 |
| 148 | Performance Issues to Consider When Migrating Production Jobs from Version 5 to Version 6 of the SAS System under MVS | Amitava Ghosh, Leigh A. Ihnen |
This paper deals with performance issues when migrating production jobs from Version 5 to Ver sion 6 of the SAS System under MVS. Due to the significant differences between the archi tectures of Version 6 and Version 5 of the SAS System, the user might experience performance degradation/improvement when migrating some jobs to Version 6 of the SAS System. Spe cific areas to consider are formats and informats, macro processing, SAS DATA STEP, and so on. This paper will try to highlight some of the known performance problem and improvement areas. It also suggests possible workarounds or altemate |
dataset, proc step, macro variable, proc means, symbol table, &variable, macro facility | source | 5 |
| 149 | Performance of the SAS System for Personal Computers under Popular Operating Environments | Allan D. Rosen | Improvements to operating environments for the personal computer have introduced credible options for SAS users. These, choices include SAS for DOS and three multitasking possibilities: SAS for osrt and SAS for DOS running under either OSI2 or Microsoft WindoWs. Performance of the SAS System in each of the four environments are compared. Factors which may effect the performance of SAS, such as disk caches, expanded and total memory, and the file system, are examined. Benchmark tests also survey the ability of a SAS user to run more than one SAS session at a time in each of the multitasking environments | regression analysis, data step | source | 6 |
| 150 | Planning, Preparing, Documenting, and Referencing SAS Products | RoJeanne Liu, Arthur Foreman |
This guide discusses quaUy control and documentation methods individuals can use to plan, prepare, document and review SAS products. It is intended for audit and program evaluation applications. The environment used here in this discussion is the U.S. General Accounting Office (GAO). The GAO is an independent, nonpartisan agency which assists Congress in congressional oversight of the executive branch of the federal government | proc step, dataset, sas log, missing value, proc print, nominal variable, data step, proc freq, if-then-else | source | 6 |
| 151 | Power Pull-Down Menus | Roger Staum | Pull-down me!lUS are a key compone!lt in. creating ·user-frie!l<ily". in.teractive SASe software applicatio!ls. This paper will help you become a· more effective pme!lU developer by showin.g you how to: impleme!lt ·pseudo coimnands·, .... hich . suppleme!lt or replace regular SAS comma!lds. These in.clude, but are !lot limited to, custom comman.ds. assigIl pme!lUS to a varieWofW'in.dOW'S with maximum fi"",biliW; This Process is not as straightfor....aro as you might assume | proc step, macro variable | source | 6 |
| 152 | Practical Econometric Analysis for Assessment of Real Property Using the SAS System on Personal Computers | Shah Amal, Robert Weselowski |
This study investigates the possibilities and potential of applying econometric (hedonic price) analysis for mass appraisal of Real property using the PC SAS System. There has been a tendency in the last decade by Real Property Taxation organizations to custom build or subscribe to customized mainframe modeling software. These systems are costly and are proving to be increasingly· restrictive with the rapid changes in computer hardware and software technology. This research focuses on the oapability of the PC SAS System in handling and computing over 53,000 records (residential property sales) wnh various property characteristics. The practical application of econometric modets along with the relevant regression diagnostics as it relates to mass property valuation is also highlighted. This paper clearly demonstrates the power, speed,reliability, flexibility and cost efficiency of the PC SAS System as a Real property valuation and research tool | regression analysis, predictor variable, condition index, sample, outlier | source | 6 |
| 153 | Predicting Charitable Behavior: A Factor Analysis of a Gallup Survey | Christopher Toppe, Arthur Kirsch, Stephen M. Noga |
The procedures available in SASISTAr' make perfonming complex statistical analyses no more difficuh than perform ing simple statistical analyses in base SAff'. As a matter of fact. PROC FACTOR. PROC STEPDISC and PROC DISCRIM can be easier to code than PROC TABULATE. Because h is easy to use SAS/STAT, there is a tendency to jump into solving a problem by doing what is called fishing. You just throw all your data into the CPU pond and see what fish you catch. H one procedure doesn' catch the right fish, you can always try another one. But SAS/STAT was never meant to be used this way. There is no better place than SAS/STAT to remember that SAS system procedures are tools, each whh a proper use. In sphe of their ease of use, the procedures in SASISTAT do not offer a substttute for careful planning and preparation before being invoked. This paper presents the process of planning and preparing for the complex analysis of a large national public opinion survey, and presents the resuhs |
factor analysis, proc step, sas base | source | 6 |
| 154 | PROC REPORT: An Introduction to the Batch Language | Ray Pass | proc report, proc step, sample, proc print, character variable | source | 10 | |
| 155 | Programming for Job Security: Tips and Techniques to Maximize Your Indispensability | Arthur L. Carpenter | A great deal has been said about programming techniques for efficiency and maintainability of SAS programs. We are taught to write code that minimizes machine and programmer resources. Unfortunately. easily maintained code requires fewer programmers. and fewer programmers means pink slips and less job security. In these troubled times, programmers need to be able to maximize their indispensability. Programmers can increase their job security and thereby protect themselves and their families by applying the tips a'nd techniques discussed within this paper. The programmer will be advised when to apply the techniques. and whether the use of the techniques should be subtle or gross. Techniques will cover programming style. editing style. statements to use and avoid. and naming conventions. You will learn to blur data steps, make non-assigning assignment statements, and in general write code that not even you will be able to figure out how or why it works | job security, data step, drop statement, macro definition | source | 4 |
| 156 | Psychometric and SAS Programming Aspects of Objective Examinations | W. J. Klein, Gordon Emslie |
Although objective achievement tests lend themselves to machine scoring, educators and programmers have not fully exploited the possibilities. Alternative methods of presenting and scoring objective tests are reviewed and those worthy of further research are identHied. Reliability, the reproducibility of measurements when testing conditions remain constant, is a prime requirement. In general, the longer the test and the less prone it is to random guessing, the more reliable. For objective tests guessing is related to the number of alternative answers. Because testing time is limited, a trade-off is necessary between a long test with few alternative answers per question and a test with fewer questions but more alternatives. Commonly, tests require the student to 'choose the single best answer provided'. However, it is demonstrated that the probability of a correct guess is typically lower when the test instructions require more than one answer to be selected | proc step, macro definition | source | 5 |
| 157 | Publication Quality Output From the SAS System | Mary A. McDonald | The advent of Hewlett Packard Laserjet and Adobe Postscript printers has created a demand for good looking output reports. It has been difficult for mainframe programmers to satisfy user expectations because of the size of files ,involved and the poor quality of high volume laser printers. This poster presents an example of a 6,000 page, publication quality report produced by imbedding printer control codes in the output file. The codes will set margins, orientation, change fonts and shade areas of the page for emphasis. The output files are created using SAS PUT statements in a _NULL_ DATA STEP. They are printed on a24 pages per minute HP Laserjet emulation printer. This is used as photo-ready copy for reproduction of the publication. The technique can also be used with any other printer control codes. Once the report is designed, the codes in one printer control language can simply be replaced by the equivalent code in another |
file statement, put statement | source | 6 |
| 158 | Putting the Pieces Together: Applying Advanced SAS Macro Techniques to Everyday Problems | Jim C. Knoop | This tutorial illustrates how SAS Macro features can drastically reduce the amount of maintenance needed by a macro based reporting system. It also illustrates techniques which help prevent types of everyday problems encountered in a multi-programmer, shared-code environment. More emphasis is placed on the creative use of SAS macros than on their syntax | macro definition, macro variable, %if-%then-%else, macro loop, macro call, raw data, sample, dataset | source | 7 |
| 159 | Putting Yourself on the Map with the GMAP Procedure | Michael L. Davis | Many users are fascinated by maps although they seem too complex to be produced by an average user. This paper introduces to new SAS/GRAPH users the concepts necessary to create maps. The paper offers gUidance in finding those pesky coordinates required to put the GMAP procedure to work. The major features of PROC GMAP that are required to draw a geographic map are explored. The paper also discussed elements of good graphic design as they apply to maps, and presents tips on using graphic devices and drivers intelligently. Part 1 |
proc gmap, proc step | source | 9 |
| 160 | Re-engineering the Drug Development Process with 'the SAS Information Delivery System | Bryan K. Beverly | The purpose of this paper is to describe how: (1) data processing points of failure interface with the drug development process, (2) the systemic causes of the points of failure and (3) to suggest how the process can be re-engineered via the SAS Information Delivery System. In the drug development process, there are several data processing points of failure; a point of failure (PDF) is a place at which problems could arise in the collection, analysis and reporting of clinical trial data. As shown by Figure 1, in moving from drug discovery to the New Drug Application, problems such as data recording (PDF 1), data interpretation, verification, validation (PDF 2), data management (PDF 3), data analysis (PDF 4) and reporting errors (PDF 5) retard the efficiency and effectiveness of the process. Figure 2 lists the typical causes for these data problems | clinical research, data science | source | 5 |
| 161 | Recruiting, Interviewing, and Hiring a SAS Programming Staff | Judith H. Mopsik | sas programming, sas programmer | source | 7 | |
| 162 | Relating Orthogonal Contrasts to Regression Coefficients through Estimable Functions in the GLM Procedure | Louis Munyakazi, Roy H. Sorbet, Richard L. Hintz |
Experiments which measure response to quantitative levels of a treatment have been widely used in clinical and pre-clinical trials. When the response to treatment levels can be fit by either a linear or a quadratic trend, the relationship between regression analysis and the results of analysis of dose response through orthogonal contrasts provides a better interpretation than either analysis separately. This paper demonstrates that by exploiting the concept of estimable functions both tasks can be completed easily and conveniently in one run of the SAS/STAT GLM procedure |
regression analysis, null hypothesis, proc genmod | source | 4 |
| 163 | Relational Processing of Tape Databases | Howard Levine | This paper covers the following topics: Explanation of Relational Processing Simple Relational Processing Why Use Tapes? Setting Up the Files Parallel Processing General Joins with More than 2 Files Limitations of Tapes | by group processing, parallel processing, data integrity, full join, inner join, proc sql, data merging, proc step | source | 9 |
| 164 | Reliability of Biostatistical Data: A SAS Macro for the Calculation and Presentation of Some Commonly Employed Statistics | Howard M. Proskin, Albert Kingman, William C. Murphy |
In the field of biostatistics, the reliability of a measurement technique plays a key role in its ability to yield data from which useful inferences can be derived. For example, in the area of bioequivalence testing, a low level Of reliability can serve to preclude the ability to identify truly equivalent agents without resorting to studies employing very large sample sizes. The quantification of reliability is a valuable first step in understanding the characteristics of a measurement | correlation, sample, intraclass correlation, macro definition, cohen's kappa, sample size, anova | source | 10 |
| 165 | Replacement of User-Supplied Dates with Variables That Were Calculated Using SAS Date Functions | Darlene M. Gerber, Allan M. Kominek |
This paper takes you through the process of replacing user supplied dates with variables that have been assigned SAS date values. It explains the code that calculates the correct SAS date values utilizing SAS date functions in Release 6.07 batch. What happens when you're assigned a project that requires a number of user supplied date changes on a monthly basis, and you, the programmer, are not going to be responsible for making these changes? The solution requires a | date variable, sysdate9 | source | 4 |
| 166 | Requests Tracking System (RTS): A Facility for the Management of Post-NDA Regulatory Requests | Robert Thomas Stemplinger | txt, accuracy | source | 6 | |
| 167 | RNs, MDs and the SAS System: Describing Medical Professional Trends Using SAS Procedures | Joseph E. Earley, Joyce Bailey, Archie J. Calise |
proc step, regression analysis, proc reg, proc univariate | source | 3 | |
| 168 | Robust Estimation Using IRLS Procedures | Eric Yu I. Wang | A SAS program, Robust, has been developed to compute parameter estimates from an Interactively Reweight Least Square procedure. This program uses a WLS reression from the PROC REG procedure, the case weight is calculated by Huba's function (C=I.345) or Tukey's biweight function (C=4.6845). The macro procedure is shown to be very nenble and allows user input data set, run bivariate or multiple robust regression model. In addition, for comparison, the OLS estimates, R2, adjuSted R2, and total sample siZe are also listed in the output |
regression analysis, proc robustreg, macro definition | source | 3 |
| 169 | Sample Size versus Detection Probabilities of Off-Aim Drifts in Support or Quality Improvement: A SAS/GRAPH Application | D. Mark Carpenter, Cecil R. Hallum |
sample, sample size, degrees of freedom | source | 5 | |
| 170 | SAS Applications Programming with DB2/SQL | David J. Adamson | dataset, proc step, variable, batch processing, select | source | 6 | |
| 171 | SAS Data Dictionary Tool Box | Allan E. Williams, Peter Dueber, Lynette M. Browne |
The need for automated data dictionary tools to document a SAS dataset is a recurring topic. This paper presents a pair of tools that link the SAS internal data dictionary information with an associated SAS Format Library to produce a SAS data dictionary report. In addition, frequency distributions for discrete variable values and univariate statistics for continuous variables can be added. Options are included to easily customize the output depending on user needs. The availability of easy to use dataset documentation tools encourages users to create variable labels and formats which adequately describe the dataset | data dictionary, numeric variable, proc step, length statement, proc contents, proc freq, proc format, interval variable, format statement, missing value, proc means, label statement, univariate analysis, dataset, sample, standard deviation | source | 5 |
| 172 | SAS Documentation: My Achy Breaky Back | Deborah Blank, Johnny Williams |
This paper discusses the ongoing commitment by SAS Institute to address users' concerns about the volume and accessibility of SASe documentation. In this paper, we showcase current and planned activities to help users of the system find the information they need quickly and easily | sas base, macro facility | source | 5 |
| 173 | SAS Macro for Analyzing Step and Impulse Response Functions Derived from Transfer Function Models | Martha Elva Ramirez-Guzman | The i :T;oul ;:·e statistics such as: the :~esoonse (.o,;:?i9i":;;s~ the si;(~'p the :tnDL~.t ~"·9S:·08nse \r~e ights (1::'- tQ-weights) used to fo~ecast future v.;lL~e·:; of the O!).t;:;u.t ser-ie5 f~~(J8 2. t:r'ansfet~ ~:~;:-;ct iC:l mode.~ ~ | source | 6 | |
| 174 | SAS Macros for Converting between ARC/INFO Single-Variable Files and SAS Data Sets | Barbara L. Jackson | dataset, macro definition | source | 2 | |
| 175 | SAS Macros: What'sReally Happening? | Mark Bercov | The SAS Macro Facility is an extremely powerful tool, which is often not utilized because it is not understood. This Hands-On PC Workshop is designed to build on the knowledge base established in previous introductory Macro Workshops and Courses. Presented in the fonn of an interactive tutorial, this workshop will delve deeper into the labyrinths of the Macro Processor to help the user understand what is really going on. However, the potential attendee can rest assured that the most innermost Macro secrets will not be revealed. Some prior experience with, SAS pata step p(ogramming and Display Manager are assumed. At least 6 months experience with the Macro Facility is highly desirable, and several unresolved Macro problems are absolutely mandatory. * | macro variable, macro facility, macro definition, macro quoting function, stored macro, flat file, call symput, error message, &variable | source | 5 |
| 176 | SAS Program Function Key Remapping In the UNIX Environment | Bruce F. Gilsen | proc step, macro definition | source | 10 | |
| 177 | SAS Software to Fit the Generalized Linear Model | Gordon Johnston | In recent years, the class of generalized linear mod els has gained popularity as a statistical modeling tool. This popularity is due in part to the flexibility of generalized linear models in addressing a variety of statistical problems and to the availability of software to fit the models. The SAS system provides two new tools mat fit generalized linear models. The GEN MOD procedure in SAS/STAT software is available in release 6.09 of the SAS system and in experimen tal form in release 6.08. SAS/INSIGHT software provides a generalized linear modeling capability in release 6.08. This paper introduces generalized lin ear models and reviews the SAS software that fits the models |
likelihood function, proc genmod, poisson distribution, proc step, response variable, goodness of fit, probability distribution, degrees of freedom, poisson regression, regression analysis, predictor variable, proc glm, maximum likelihood estimation, estimate, main effect, interval variable, binary variable, insurance claims, binomial distribution | source | 8 |
| 178 | SAS System Graphical User Interface (GUI) Changes and Enhancements | Claire Cates | This paper describes the graphical and text user in terface of the SAS System and how the SAS Display Manager takes advantage of the interface. The topics discussed in this paper include: how windowing environments improve pro- grammer productivity the evolution of the SAS windowing system the base windows in display manager the software layering that is used to achieve portability * the future directions for display manager and the windowing system | sas output, sas log | source | 9 |
| 179 | SAS System Visual Performance in Evaluating Arrhythmia Time Series | B.J. Cerimele | Visualization techniques are investigated as an adjunct to analytical methods in discerning and characterizing the dynamical aspects of arrhythmia time series. Performance of the SAS graphics packages, SAS/GRAPH, SAS/INSIGH-rBl, SAS/DATASCOPE, and SAS/NVISION@, is scrutinized in melding visual acumen with statistical analysis for distilling the informational content of time series data. The data were collected in a prospective cohort study of 60 healthy adult males undertaken to characterize the spontaneous incidence and temporal aspects of ventricular arrhythmias. Ambulatory circadian recordings of the electrocardiogram, through Holter monitoring targeted for 42 consecutive days, were reduced to hourly rates of ventricular premature beats. These series were assessed for evidence of cyclical, chaotic, and noise content and for the occurrence of clustering |
time series analysis, proc sgplot | source | 3 |
| 180 | SAS Transport Files or SAS/CONNECT Software: Considerations for the Proper Choice | Tamara R. Fischell | In an ever-growing multivendor architecture in and out of the workplace, an increasing need grows to move SAS datasets across multiple operating sys tems. To move SAS data across different, incompati ble operating systems, one must either use SAS/ CONNECT software or create SAS transport files. The decision to use SAS/CONNECT software (merely as a means of transfer) or to create transport files is a serious matter. In an attempt to make the choices more clear, this paper explores several con siderations. The advantages and disadvantages of using transport files and SAS/CONNECT software are explored and compared according to certain pri orities. In particular, this paper focuses on the fol lowing issues: Cost Efficiency and Flexibility * Availability |
dataset, sas base | source | 8 |
| 181 | SAS/AF Software Information Systems on Multiple Platforms: Welcome to the Real World | Burtt Blodgett, John Powers |
The Load Data AnalysiS Workstation (LDAW) is a SAS/AF app6cation developed by Quantum Consulting Inc. for the Electric Power Research Institute (EPRI). LDAW is designed to accass. analyze, and present data collected from electric utility load research programs. LOAW serves as an intuitive front-end to more than 100 SAS programs. Atthough originally designed to operate within a PC environment, SAS transport facilities and SAS Multi·Engine Architecture have allowed the LDAW to be successfully ported to UNIX and MVS with only a modest amount of effort. However, managing multi-platform development and support is a difficult and at times overwhelming task. This paper addre~s many issues involved in multi~platfonn development, and w,lIreveal SASICONNECT as a solution for many multi· platform development and network. connectivity issues |
proc step, proc cport, dataset, proc download, proc upload | source | 8 |
| 182 | SAS/EIS Software: A Tutorial | Scott Isaacs, Gail Kramer |
dataset, bar chart | source | 7 | |
| 183 | SAS/ENGLISH Software: A User's Story | Patrick J. Keating, Kevin M. Rapp, James V. Gigliotti, H. John Maxwell, Scott R. Hampton |
SASIENGLISH software alloYfs you to ask questions about your data in plain English and view the answers to your questions in a variety of ways. Building a SASlENGLISH application requires you to create a Knowledge Base that describes the data. This article recounts our experiences in setting up a Knowledge Base for a SASJENGLISH application to aid the Managed Care Programs of Pennsylvania Blue Shield | sample, permanent library | source | 5 |
| 184 | SAS/LAB Software for Guided Data Analysis | Wayne E. Watson, Katherine J. Roggenkamp, Randall D. Tobias |
SAS/LAB software is a data analysis advisor for engineers and scientists who lack an extensive statis ti~al b\\ckgrpl!nd. The spftware is implemented ~ a Inentt system, making it'very easY,tcdearnand use. As an. analysis proceeds, the software explains and in terprets~he results, makes suggestions for further anal ysis, and -issues warnings about violated assumptions. It also provides convenient methods for converting raw data to a usable form, producing graphical display of |
data visualization, dataset, character variable, anova, scatter plot, outlier, homoscedasticity, main effect, regression analysis, data science | source | 7 |
| 185 | SAS/PH-Clinical Software: CANDA Implementation Strategies | Juliana M. Ma, Sandra D. Schlotzhauer |
SAS/PH-Clinicar" software is currently used for a variety of clinical data review needs w~hin the pharmaceutical and biotechnology industries. Certainly, one of the most critical needs for effective delivery of clinical information comes when implementing a Computer-assisted New Drug Application (CANDA). SAS/PH-Clinical software is an obvious choice to provide the data function of a CANDA system. The intent of this paper is to provide tips and techniques for utilizing SAS/PH-Clinical software for a CANDA situation. Strategies for developing a successful CANDA system based on SAS/PH-Clinical software are discussed. Emphasis is placed on the signfficance of effective coordination w~h the traditional in-house review effort and the benefits of applying that knowledge when setting up a computerized system for regulatory agency use. Given the importance of planning, future enhancements to SAS/PH-Clinical software that would benef~ CANDA systems are highlighted |
clinical trial, clinical research, phase ii, data science | source | 8 |
| 186 | Secrets of a Successful SAS Users Group | Sally A. Roberson | source | 4 | ||
| 187 | SET, MERGE, UPDATE, or MODIFY? (And What about Indexing?) | Phil J. Weiss | modify statement, set statement, by group processing, update statement, dataset, master data management, sas base, index creation | source | 6 | |
| 188 | Sheet Product Conformance Reporting System | L. Richard Woodyatt, John A. McNamara |
A menu-driven, SAS-based system has been developed to generate the product specification conformance reports that are now required by many customers of tin-plated and cold-roJled steel products. This system, which was designed for use by the Quality Assurance Department, also serves a dual function of guiding the quality control engineers in taking any required corrective actions that may be necessary to maximize product capability. It provides authorized users with SAS/AF and SASIFSP screens to define the nature arid contents of each customer's reports, the properties to· be inCluded, their control limits, and time period(s) to be covered by the reports. Based on these report characteristics, the SAS system then build~ a custom SQl query to extract the required data from the various DB2 foundation data systems |
standard deviation, sample, histogram, dupout= | source | 8 |
| 189 | SIMSPRT: A SAS Code for Simulation of Sequential Probability Ratio Tests | Kenny C. Gross, Kristin K. Hoyer |
Expert systems are being developed at Argonne National Laboratory for signal validation and equipment operability surveillance in a variety of reactor and industrial applications. These systems are based upon the Sequential Probability Ratio Test (SPRT), a sensitive sequential test which provides early annunciatio~ of the the onset of degradation in sensors or eqwpment SIMSPRT is a SAS" code used as a tool in the design of and reliability analysis of SPRT modules for new signal applications. Utilizing Monte Carlo simulation, SIMSPRT estimates the false-alarm and missed-alarm probabilities for an individual SPRT module as a function of the input signal noise distribution. In addition to reliability analyses, SIMSPRT can be used to simulate a variety of known sensor degradation events for purposes of determining the sensitivity characteristics of a SPRT module and the maximum time-to-annunciation for subtle degradation modes. I | monte carlo, proc freq, data cleaning | source | 6 |
| 190 | Simulation Models for Business Planning and Economic Forecasting | Donald Erdman | proc model, proc step, monte carlo, covariance, estimate, proc nlin | source | 7 | |
| 191 | Software Intelligence: Applications That Customize Themselves | LeRoy Bessler | horizontal bar chart, bar chart | source | 9 | |
| 192 | Some Recommendations for the Selection of a Response Surface Design | Douglas C. Montgomery, James E. Taggart |
source | 8 | ||
| 193 | Some Unusual SCL Techniques | Paulette W. Staum | A variety of techniques can i1crease your effectiveness and effICiency when you use Screen Control language (SCl). These techniques include: 1) using macros, program entries or method blocks as utilities 2) using templates for common types of . programs 3) using SUBMIT blocks for convenience of coding or speed of execution 4) executing commands for a user 5) using SCl in batch jobs Examples include: 1) a utilitY to count the number of observations currently available from a data set with a WHERE applied 2) creating a selection list based on a format 3) quicker input/output by using SUBMIT btcicks 4) an automated procedure to add SCl programs to a catalog of FSEDIT screens 5) a utirlty to identify the fibrary contaning a format INTROIlJCTION Screen Control language (SCU is a powerful tool for writing user-friendly interfaces. As a developer, you want to develop an interface that is user-friendly, consistent, and predictable | proc step, batch job, macro variable, sas base, proc format | source | 6 |
| 194 | Spectral Decomposition and Reconstruction of Nuclear Plant Signals | Kristin K. Hoyer, Kenny C. Gross |
A reactor parameter simulation system (RPSS) has been developed in SAS" with the pwpose of analyzing serially· correlated signals, decomposing those signais into deterministic and stochastic components, and reconstructing new signals that possess the same correlation structure as the original signals. The RPSS provides (I) a simulation tool for plant control strategy development, and for safety study investigations of transient events (such as a degraded sensor) and (2) a vehicle for filtering serial correlation from reactor signals, which is important in our ongoing development of expert systems for sensor-operability surveillance. The power and utility of theRPSS is demonstrated on signals from ANL's EBR-II nuclear reactor and verified through various statistical tests. I. Introduction: A Reactor Parameter Simulation System (RPSS) bas been developed in SAS'" at Argorme National Laboratory (ANL) in Argorme, illinois | correlation, macro definition, raw data, proc freq, proc step, histogram, data cleaning | source | 6 |
| 195 | STAB: A PC/SAS Stability Analysis Software | Moh-Jee C. Ng, Yi Tsong |
data science, scatter plot | source | 8 | |
| 196 | Statistical Graphics in Environmental Regulatory Studies | Larry R. Holden | Agricultural companies are required to conduct large numbers of environmental studies to evaluate the potential impact of their products on the environment. The | sample, proc calis, sample size, histogram | source | 8 |
| 197 | Statistical Procedures in Automated Cervical Cytology | David L. Sherry | sample, sample size, type ii error, null hypothesis | source | 4 | |
| 198 | Strategies for Performing Multiple Comparisons on Means | Jenny Aquino Kendall | This Morial descrlbes four strategies for perfonning multiple comparisons on means using PROC ANOVA and PROC GLM. Example data are presemed and analyzed w~h a two-way factorial treatrnem model. Multiple comparisons on the imeraction means are obtained . and the results compared using the methodologies of (1) reformulating the model to a one-way ANOVA, (2) using BY grouping, (3) investigating LSMEAf\lS, and (4) performing CONTRASTS |
anova, proc step, proc glm, proc genmod, main effect, proc anova, mse, least squares mean, significance level, proc sort, by group processing | source | 7 |
| 199 | Strengths and Weaknesses of the SQL Procedure Compared to Standard SAS Language | Howard Levine | proc step, proc sql, format statement, data step, create table, proc print, proc sort | source | 8 | |
| 200 | Structured Systems Engineering and the SAS System: Godzilla Versus Megalon | James E. Johnstone, Stuart Leven |
The SAS system is well known as a powerful data analysis and Iq>OIt genOIlllion system, and its capabilities .... now extending into Bxocutive Infoanatioo Systems (EIS) and other enleIprise wide infoIlllation systems applications. It is also a fuJI-fealuled programming language suitabJe for use in baditional, large-scale, production-level data processing systems. 'Ibis paper examines the relationship, sometimes harmonious and sometimes no~ between the SAS system and the s1IUctured systems engineering techniques used to build baditional systems. To the extent that this .xelationship is not harmonious. the authors suggest ways to oven:ome or mitigate the di1liculties | dataset, proc step, macro facility, macro variable, data science, macro definition, data wrangling, sample | source | 10 |
| 201 | Support for Sales Figures Analyses with VIPAS at Deutsche Bank Life Insurance | Berthold Kerl, Wolfgang Scholz |
life insurance, broker | source | 14 | |
| 202 | System Trending for Statistical Analysis of System Performance Measurement of Plant Wellness | Joseph G. Moyer, Steve Leshnoff, Mark Orski, Judy Keiper |
Statistical analysis of in-service data is a suitable means to numerically characterize process trends and provides the quantitative basis for predictive preventive plant maintenance. When on-line instru mentation is available, the high numerical resolution that is achieved by the collection of plant transient data, combined wnh the process analysis and statis tical tracking of quickly changing events, provides a high confidence that nothing will escape detection and quantification. Equally important is the utilny of modeling historical su rveillance data in anticipation of potential process trends as performance indica tors of system deterioration. Paramount to the success of this effort is the availabilny of the SAS procedures and interactive components | proc step, data science | source | 6 |
| 203 | Techniques for Merging Non-Indexed Data Sets with Duplicate Values of the BY Variables | Mark Prior | By using the MERGE statament. SAS Systam users may perform one~to-one and many-ta-one data set merges. However, the MERGE statement is incapable of accurately executing many to-many data set merges. This paper K1entifies four routines capable of accurately performing many·to·many (duplicata values of the BY variables) data set merges. All of the routines were written in SAS/BASE (version 6.07) and tes\ad on an IBM 3090 mainframe computer under the MVS/ESA operating system. Each of the routines applied a different strategy to merge the input data sets. One method utilized the data set transforming capabilities of PROC TRANSPOSE, while another used PROC SOL to perform the merge. A third methodaocomplished the merge by controlling the input data set pointers and performing a binary search. A fourth technique accomplished the merge, by using data set pointer control, an initial binary search and a proximity search |
cpu time, proc step, merge statement, proc sql, proc transpose, pointer control, do loop | source | 6 |
| 204 | Techniques for Sharing Screen Control Language Programs ih a SAS/AF Application | Derek Drummond, Veronica R. Walgamotte, Jeffrey Phillips |
sas library, conditional do loop, sas base | source | 5 | |
| 205 | The 20-Minute Installer-Using a Laptop to Distribute the SAS System for PCs | John Zemba | This paper describes a method of distributing the SAS System for PCs in about 20 minutes for each workstation. My objective as the SAS/PC consultant was to minimize installation related problems while ensuring that all users had similar installations. Rather than bury the users with quick reference sheets full of installation and testing guidelines, I decided that with this method I could perform the entire installatio'nmyselfin about 20 minutes by using a laptop computer and a high-speed file transfer utility called FastLynxTM. The approach I developed allowed me to installSAS/PC on one computer, test all of the individual SAS modules, and quickly transfer the software to another machine. Variations in hardware/software configurations, diskette sizes, and skill levels were no longer considerations. Installation I installed a copy of the SAS system on a Compaq SLT /286 computer. The modules we licensed consisted of BASE, |
sas consultant, sample | source | 3 |
| 206 | The Evolving Role of SQL in Genetics, Breeding, and Selection Work | Carmela C. Janagap, Ambekar A. Eknath |
The Genetics Improvement of Farmed Tilapias (GIFT) project utilizes procedure SOL of the SAS System 10 manage, store and cross-correlate multiple sets of data generated from successive generations of selective breeding. Practical queries developed in SASISOL" will be presented and compared with 1radilional DATA and PROC steps. Impfieations 01 using S1ruc:IUre<i Query Language to the current selection and breeding work of the project will be discussed |
proc step, proc sort, order by, proc sql | source | 5 |
| 207 | The FlLEVAR Option: A Tool for Management of Large Numbers of Raw Data Files | Henry W. Buffum, Jeffrey S. Rosen, Richard W. Latimer |
When building a SAS database system, it is usually preferable to store all data in SAS datasets. However, in many scientific databases it is necessary to store some data in ascii data files. Files produced by laboratory instruments or monitoring devices may need to be kept online in their original raw format. And scientists may prefer to use their own FORTRAN programs or other software to access these raw files. Fortunately, it is not difficuH to design a SAS database system that can manage raw ascii data files. A new feature of the IN FilE statement, the FllEVAR option, is particularly useful in managing large numbers of raw data files. This paper describes a SAS-based environmental database system that allows easy access to large amounts of data stored in raw ascii format | dataset, data step, infile statement, file statement, raw data, input statement, end= option, proc means, proc step | source | 4 |
| 208 | The Generation of a Series of Annotated and Templated Statistical Graphs Using SAS Macros | Susan J. Kenny | This paper describes the strategy used to automate the production of a series of annotated and templated statistical graphs. Seven macros, written with the SAS software, are used in this approach. The first macro, %MINMAX, determines the minimum and maximum values for the X and Y axes over the range of plot coordinates. Using these values, appropriate sizes for the axes increments are determined. These macro variables are passed to two macros, %PLOT1 and %PLOT2, that produce X Y plots for the macro specified variables of &X and &Y. The %ANNOmacro utilizes an output data set from PROC MEANS to place descriptive statistics within a template to the side of the graphic display. Two. add~ional macros, %TITLE and %DESCR are used to produce the graph t~le and additional annotated descriptive text. The % TREPLAY macro places the graphic streams into the appropriate template panels and produces the hard copy output |
macro variable, data visualization, predictor variable, macro definition | source | 5 |
| 209 | The Importance of SAS Software in a TQM Environment | Paul J. Waldowski | In today's dynamic business environments, top management leaders are initiating the integrated management system known as Total Quality Management (TQM). This strategic system for achieving customer satisfaction through employee empowerment and teamwork by using software like SAse to continuously improve an organization's processes will ultimately lead to quality. The major emphasis of this paper is to expose some of the outstanding quality facets of SAse software an organization can exploit to meet its customers requirements, needs, and most importantly, their expectations. References to SAS· software in this paper are for illustrative purposes only and do not constitute an endorsement of the product by the U. S. Air Force or the Department of Defense | customer satisfaction, quality assurance | source | 5 |
| 210 | The Laboratory Review System: A SAS-based Graphical and Statistical Tool Developed for a Computer-Assisted New Drug Application (CANDA) Submission | Russell R. Holmes, David A. Bright |
The Laboratory Review System is a user-friendly graphical and statistical tool developed primarily for the review of laboratory data. This SAS-based information system is a part of a much larger system that is presented to the Food and Drug Administration in the form of a Computer-Assisted New Drug Application (CANDA) submission. The Laboratory ReviewSystem utilizes a wide variety of innovative uses of SAS Institute software products. The primary strengths of the system are the custom graphical representation of laboratory data generated from clinical trials, the identification and 'drill-down' mechanisms for patient outlier information, and the ability to perform ad-hoc descriptive statistics across many clinical variables. The major graphical component of the Laboratory Review System is a custom, data-driven algorithm which produces symmetrical scatter plots of patients pretreatment (baseline) values versus posttreatment laboratory values | time series analysis, laboratory data, proc means, proc calis, series plot, outlier | source | 6 |
| 211 | The MODECLUS Procedure: Nonparametric Significance Test, for Clusters | Ann Kuo | The MODECLUS procedure, available in release 6.08 of SAS/STAT software, clusters the observations in a SAS data set using any of several algorrthms based on non· parametric densrty estimates. The data can be numeric coordinates or distances. PROC MODECLUS can perform approximate significance tests for the number of clusters and can hierarchically join nonsignificant clusters. This paper gives a brief overview of the densrty estima· tion, clustering methods, and significance tests in PROC MODECLUS. How PROC MODECLUS differs from other clustering methods is explained, and simulations resuks for sample sizes ranging from 20 to 2000 are shown |
sample, sample size, normal distribution, proc step | source | 6 |
| 212 | The One-Eyed Guide to Internet, EMAIL, LlSTSERV, and SAS-L | Michael L. Davis | One of computing's ironies is that sophisticated users are often "modem illiterate." The modem illiterates never enjoy trading SAS" tips and "flames" outside of conferences. The ability to exchange ready-to-run SAS code and messages instantly but courteously is denied to them. They can seek technical assistance only when someone is available to speak to them. In an effort to stamp out this heinous form of illiteracy, the One-Eyed GUide is offered | source | 10 | |
| 213 | The Pricing of Stock Returns That Have a Time Varying Beta Coefficient Using the SAS/IML Module | Michael Murray | Generally speaking, regression and time se ries models are considered to be station arYi that is the same fixed model structure is assumed to hold for all observation pe riods. Furthermore, these models gener ally do not provide a facility for incorpo rating any subjective information that one might have. Clearly, given a financial mar ket setting which is continually responding to changing political, social and economic conditions, such assumptions are both re strictive and unrealistic in nature. Harri son and Stevens (1976) have developed a class of state space models that are both flexible enough to allow for the incorpora tion of subjective information, via the ad justment of suitably chosen parameter val ues, as well as time varying in nature in that a single probability model structure is not always assumed to exist for all observa tion periods | covariance, maximum likelihood estimation | source | 7 |
| 214 | The Progression and Integration of the CANDA at Merck | Bruce Binkowitz, Wanda Bidlack, Nancy Bauer |
Merck has submitted their second generation Statisti cal Computer Assisted New Drug Applicatioll (CANDA) to the Food and Drug Administration (FDA). Continued development at Merck of the CANDA focused on the implementation of advanced features, analysis and imple mentation of a new hardware platform and the initial incorporation of the CANDA as an internal New Drug Application (NDA) development tool | clinical trial, data science, data cleaning, proc step, dataset | source | 6 |
| 215 | The SAS System for OpenVMS Migrating from VAX to RISC Architecture | Ken D. Worsham | The availability of the OpenVMS RiSe hardware piaUorm and OpenVMS means users have increased price/performance available for using the SAS System for VMS. This paper discusses technical differences between the RiSe and VAX implementations of OpenVMS that will affect VMS users involving system configuration and tuning for use of the SAS System. It also discusses operational Issues of sharing SAS data and applications between trad_ional VAX and RiSe Implementations of OpenVMS | dataset, variable, decimal, missing value | source | 5 |
| 216 | The SAS System in a Longitudinal Environmental Health Intervention Study with Repeated Monitoring | WinKey Pan | In this paper, we will focus on how SAS has been applied in managing, analyzing data, and preparing reports for a longitudinal neighborhood-wide environmental health intervention project. Special attention will be given to the following topics. 1. How PC SAS, mainframe SAS, and xBase interact with each other in data management. 2. How SAS Macro Language has been used in repeating similar data analysis and generating stepwise data modeling procedures, e.g., stepwise GLM, | macro definition, data science, macro facility, sample, proc sgplot, proc means, imputation, proc genmod | source | 4 |
| 217 | The SET Statement and Beyond: Uses and Abuses of the SET Statement | S. David Riba | The SET statement is one of the most frequently used statements in the S~ System. It is also probably one of the more misused SAS statements because the SET statement typically works without much programmer input. Proper use of the SET statement is one of the key techniques to improving the efficiency of SAS programs |
set statement, by group processing, where statement, dataset, _eof_, numeric variable, conditional do loop, proc step | source | 7 |
| 218 | The Urge to Merge: A Computational Method for Linking Datasets With No Unique Identifier | Joan Kessey, Robert M. Bell, Toni Richards |
This paper describes a procedure used to link California vital statistics records for. very low birth weight (VLBW) infants (weight less than 1500 grams) to Medicaid claims data. Because the two main data files share no unique identifier, record linkage required combining evidence across several identifying variables such as name, birth date, and delivery hospital. We used record linkage theory to compute scores that measure the likelihood of a true match, i.e., that two records correspond to the same delivery. These scores weight appropriately the various pieces of evidence for or against a match. The scores allow for miSSing data, data errors, and other factors that may cause a true match to disagree on SOmE! variables. The resulting procedure is effiCient, accurate, and noniterative | record linkage, proc step, proc format, sample, unique key, proc freq | source | 6 |
| 219 | The Use of Biplots in Diagnosing Interaction Patterns of Two-way Classification Data | William J. Price, Bahman Shafii |
main effect, anova, proc step | source | 6 | |
| 220 | The Use of Cluster Analysis in Molecular Design | C. Frank Shen, Kimberly A. Mason, Alan H. Katz |
In drug design, it is widely accepted that two molecules have the same biological activity if they interact with the same receptor in the same way. To identify a potential drug, for example, a chemist would fit different conformations of a candidate to the target molecule manually using the Cartesian coordinates and could obtain further information by calculating non-bonded potential energies which resulted from probing the surface around each molecule with hydrogen bond donor, hydrogen bond acceptor. and lipid probes. A statistical procedure combining hierarchical cluster analysis with principal components is developed to identify clusters that best account for similarity among the targ&t structures using the same data mentioned above. This procedure makes it possible to automate the fitting process by comparing orientation and energy of probes simultaneously. Analysis was conducted using statistical procedures in SAS" software | cluster analysis, proc princomp, proc step, hierarchical clustering | source | 7 |
| 221 | The Use of SAS Macros and Formats in an Information Delivery System | Clive Cooper, Clare Somerville |
This paper describes the development of an information delivery system written in SAS macros, using a data dictionary to generate the SAS macro calls and format statements. Ten data sources, totalling some 600 Mbytes of data, were reduced to around 2,000 data variables for use in management reporting, and as input to an expert system monitoring New Zealand health indicators | macro call, txt, data dictionary, macro definition, extract, macro variable, %if-%then-%else | source | 7 |
| 222 | The Version 6 SAS System Under VSE/ESA | Paul Mortensen, Catherine B. Corbett |
In the past few years, IBM Corporation has significandy extended and enhanced the VSE operating system. The SASe System has also grown and matured with Release 6.08. With the addition of new functionality and the removal of old limitations. Release 6.08 of the SAS System under the VSE operating system is more flexible and powerful than ever before. This paper overviews the implementation of the SAS System under VSE, especially VSElESA -. Emphasis is on those enhanoemenlS unique to the VSE operating system with some discussion of more generalized aspects of the SAS System | sas library, batch job, sas batch, sample, sas base, options statement | source | 6 |
| 223 | Tools and Techniques to Integrate and Map Geographic Data Using SAS/GRAPH Software | Jay Jacob Wind, Bernard R. Baum, Chapman P. Gleason |
proc step, proc gmap | source | 9 | |
| 224 | Tuning Tips for Applications That Use SAS/SHARE Software | Bill Brideson, Stephen M. Beatrous, Daniel J. Squillace, Jan Squillace |
The SAS System has many tuning options. most of which are left at their default values. When an application accesses data through a SAS server, sometimes the default values provide adequate performance and sometimes they do not. The SAS System is delivered to you prope~y tuned for a ··typical" application that uses SASISHARE software. SAS Institute makes some assumptions about the kind of processing that is going to take place in a "typical" application. Recognizing that your application may not be "typical" I the SAS System supplies tuning options that you can use to override default behavior. This paper discusses programming techniques and option value adjustments that you can use to improve the performance of your applications that access data through SAS servers. The information In this paper is the result of tuning several large applications that are in use at SAS Institute | source | 9 | |
| 225 | Understanding the Transpose Procedure | Susan P. Repole | proc transpose, proc step, numeric variable, character variable | source | 5 | |
| 226 | Universal Data Access: A Success Story | Bruce Babbitt, Marianna Suciu, Emily P. Wallace |
Data are a valuable asset to a corporation but,data are stored in many different forms on many different hardware platforms. Getting to all ofthe data needed to make an informed decision can be a challenge because it can require the use of several different tools on several different operating systems. By the time the data are gathered and the decision is made, the business opportunity could be lost. SAS! ACCESS~ software provides a direct and transparent interface to your database data and can be used in conjuction with other SAS* software to access all of the data in your enterprise. The same set of programming statements are used to access data in SAS mes as well as the data in databases so you use one set of tools for all of your applications. It is very easy to gather information quickly when you need to make important decisions. This paper demonstrates the ways that SAS/ACCESS products can be used to build applications that retrieve data from a variety of sources |
data science, data governance | source | 5 |
| 227 | Using GOPTION Statements for PC/SAS Graphic Devices | Randy Van Beek | source | 3 | ||
| 228 | Using SAS Software in Pharmacoepidemiologic Research: Identifying Episodes of Drug Use and Determining Average Daily Dose | Charlotte Corelle, Marjorie Bennett, Bentson McFarland, Richard E. Johnson |
This paper describes how we used base SAS software to transform pharmacy records of all dispensings of a particu lar drug to records of all episodes of use of that drug. The challenge was to develop a sound and efficient process that would correctly link dispensings into episodes. This paper presents a flexible algorithm for constructing episode records from dispensing records and displays the code for imple menting the algorithm. The algorithm uses variables com mon to many pharmacy databases (i.e., user number, prod uct number, dispensing date, dispensed quantity, strength, and day's supply) to set the bnundaries of a user's dose and duration for each period of exposure to a particular drug. The flexibility of the algorithm comes from its use of some function of either quantity andlor day's supply andlor mil ligrams of drug dispensed to decide whether or not to link one dispensing to another as part of the same drug use episode | unique key, sas base | source | 3 |
| 229 | Using SAS Software to Create a Modular Forecasting System | Tom Fischetti, Steve Heathcate, Don Perry |
Flexibility is always an important part of any database system and SASe has provided the needed tools. In the past, forecasts were loaded and analyzed in a nonstandard format using spreadsheets. These spreadsheets Wefe cumbersome, time consuming, and resulted in the forecaster spending the majority of his/her time on mechanical data processes rather than concentrating on modeling/analysis functions. Using SAS/AF, SAS/FSF', SAS/STA T, SAS/IML , SAS/ETS and a host of SAS procedures, the Usage Forecasting System (UFS) was developed to provide the ability to load, evaluate, validate, edit, and forecast over 480 items. This system is divided into four basic modules consisting of ACTUALS, MODELS, FORECASTS, and REPORTS. Actuals are loaded through ASCII files or direct data set editing. Multiple models are allowed for each of the 480 items forecasted. The multiple models for a particular combination are weighted based on past performance and then combined into a single forecast |
proc step, dataset | source | 6 |
| 230 | Using SAS Software to Develop an Online Pregnancy Registry System | Jean Fletcher | Release 6.07 of The SAS System for information delivery has extended its capabilities for data access, data management, data analysis, and data presentation. It also has enriched the environment for SAS application development This paper discusses how we integrated several SAS products, SAS/AF@, SAS/FSP@. SAS/BASE@. SAS Screen Control Language (scll, and SAS Structured Query Language (SQl) in version 6.07 to develop a user-friendly on-line information system - Pregnancy Registry System (PRS) |
proc step, create table, proc sql, proc forms, data cleaning | source | 6 |
| 231 | Using SAS Software to Identify Critical Variables in the Manufacturing of Hybrid Microelectronics Assemblies | Melvin T. Alexander, Brian Hogan, Ingrid Jensen Vaughan |
In the production of hybrid microelectronic assemblies, the wire bonding process is a large yield detractor that poses many opportunities for product failure. Technological advances are requiring a high number of wires on future hybrid assemblies; therefore, manufacturers are striving towards continuous improvement to reduce the defects from parts per thousands to parts per million or billion ranges. Statistical Design of Experiments (SDOE) is a key tool for developing, optimizing, and improving processes by distinguishing between major and minor contributors of quality characteristics. SDOE helps uncover the crttical factors and settings of a process or product and their effects on process variabilny. This paper discusses the use of SAS software to design and analyze an experiment to identify critical substrate manufacturing variables that affect the wirebonding process. Also, the paper discusses a comprehensive approach used to achieve continuous process improvement | proc step, proc genmod, proc glm, anova, regression analysis | source | 9 |
| 232 | Using SAS/ACCESS Software with PC Files and Databases | Forrest Boozer | This paper discusses the SAS/ACCESS software products available under OS/2 and Microsoft Windows and describes the way the products work in both environments. The paper explains the differences i)etween the interfaces for DIF files, DBFfiles, Database Manager, and AS/400 Data and demonstrates how to use them efficiently. Plans for future enhancements to the SAS/ACCESS products for pes are also discussed |
select, proc sql, where statement, proc step, order by, variable, date variable, time., missing value, sample, dataset, relational database | source | 8 |
| 233 | Using SAS/AF and SCL Extended Tables to Build a Custom-Designed Data Editor | Darius S. Baer | Although widely acknowledged as an industry standard for managing data and information, the SAS System has come' into its own as a sophisticated, easy-to-use application generator. Custom-designed data editors are in great demand. With the SAS Application Facility (AF) and Screen Control Language (SCl), the developer can quickly prototype enduser requirements. As the enduser previews the application and recommends changes and enhancements, the developer can rapidly-add and implement the additional code. The SAS System provides an environment in which the application requirements do not have to be totally foimalized before development is begun. In fact, all . that is needed is a general understanding of the business needs and the environment in which the application will be used. General editors differ greatly from Application Specific Editors (ASEs) | dataset, proc step | source | 10 |
| 234 | Using SAS/ETS in Applied Econometrics: Parameter Estimates for the CES-Translog Specification | John J. Hisnanick | estimate, proc step | source | 5 | |
| 235 | Using SAS/GRAPH Software to Create Unclassed Choropleth Maps | Michael Swift, Diane Nishri |
Traditionally choropleth maps of continuous data are produced by first grouping the values of the variable to be mapped into a small number of class intervals. Shading is then applied to each geographical area of the map according to the class assigned to Ifle area Unclassed choropleth maps avoid this step by applying shades whose visual intensity is directly proportional to the magnitude of the area's specific data value. The unclassed method thus _ eliminates the problem of determining the optimum number and range of class intervals, and any subjective bias associated with this process. As a result of not suffering from quantization error. unclassed maps are a more accurate representation of the statistical distribution of the data being mapped | response variable, proc gmap, proc step, outlier, sample | source | 6 |
| 236 | Using SAS/IML for Multiyear Rotation Design Sampling and Estimation in Agricultural Surveys | Yillian Yuan, Raj S. Chhikara, Lih-Yuan Deng |
We first d.escri1)e l!. proposal. by. Chhikara and Deng( 1992) which is based on an,analysisofvariance model that takes into account the successive sam pling of units in the area fr~e across years as sam pled by the U. S.Depart$entofAgriculture(USDA). We then discuss some of the computational prob lems related to the proposed estimation procedure. An implementation usingSAS/IML (1989) for the multiyear method applied to the actual data from US DAis .discussed in this paper .* | sample, anova, covariance | source | 6 |
| 237 | Using SAS/IML Software to Generate Odds Ratios and Associated Intervals Based on Logistic Regression Output | Catherine O. Cowie, Danita D. Byrd-Hol, Keith F. Rust |
Odds ratios and associated confidence intervals are frequently used measures in epidemiologic and biomedical research and are often derived from logistic regression models. Computation of these measures is straightforward when based on simple logistic regression models. Many logistic regression models, however I are complex and involve interactions of independent variables. In this case, computations of these measures may be quite tedious and subject to error if performed by hand. An automated .method using the SAS software is not currently available for use in these more complex regressions. We present a simple and quick computer program for calculating odds ratios and confidence intervals from logistic regression output using SASIIML software. The method may be used both for reporting results from a final reduced logistic regression model and for model fitting in which the magnitude of effect of a variable is a criterion for model inclusion | odds ratio, confidence interval, regression analysis, logistic regression, logistic distribution, predictor variable, covariance, proc iml, effect size, model training, main effect | source | 6 |
| 238 | Using SAS/PUBLIsH Software as a Document Creation System | Eloise Currie | The Publications Division at SAS Institute is converting atl documentation style programs to SAS/PUBLLSH'" software, the Institute's new document processing product. SAS/PUBLISH software offers a number of advantages for SAS users, as it is designed to incorporate output from other SAS products and can also be used to create formatted documents which can be viewed online. Writers have found these to be great advantages, as elements such as SASJGRAPH output can be incorporated in a document and routed to any hardcopy or softcopy PostScript output device. The elimination of hardcopy drafts and the easy inclusion of graphics have reduced writing and production time for documents. This poster will focus on: document style elements written in the SAS/PUBLISH language how SAS/PUBLISH software is meeting our need to integrate output from other SAS products and procedures how SAS/PUBLISH software fits the Institute's goals for future hardcopy and online documentation |
proc step, if-then-else | source | 4 |
| 239 | Using SCL to Build Portable UNIX Applications | Thomas C. Edds | sample, dataset | source | 8 | |
| 240 | Using the FRAMf Entry to Develop a Graphical-based Patient Laboratory Safety Review Application | David A. Bright | Screen Control Language (SCL) and SAS/AF software has enabled users to develop custom applications to take advantage of the analysis tools that the SAS System provides. A new entry within SASI AF called FRAME allows the user to combine the ability to display graphical information with the power and versatility of a SCL program screen. Graphical representations of laboratory data provide an efficacious means for analyzing overall patient safety. The FRAME entry allows the user to identify patients from pre- vs post-treatment and time series laboratory scatterplots. Utilizing novel SCL code, hot spots, and the FRAME entry allows for the "drill-down" and identification of patient outlier information. The use of this application has facilitated a more expeditious review of safety by our in-house physicians |
laboratory data, proc step, concomitant medications | source | 9 |
| 241 | Using the Release 6.07 PRINT Procedure Enhancements to Improve Your Output | Debi M. Reece | I use SAS daily to produce quick adhoc reports. For years. I have patiently answered three questions with a | proc print, proc step, by group processing, title statement | source | 6 |
| 242 | Using the SAS ADX Menu System for Continuous Quality Improvement | Melissa A. Durfee | Utilizing the SAS ADX menu system, process optimization through identification of significant variables and examination of sources of variation is realized. This system, included with SAS/QC software, facilitates design and analysis of experiments by providing a full-screen front end. For the novice user and experimenter, prompting menus are provided which guide one through building the design, printing data collection forms, and performing standard statistical analyses. However, designs may be customized by turning off the prompts, yielding a wide variety of statistical options |
main effect, standard deviation, data science | source | 6 |
| 243 | Using the SAS DATA Step Language to Read dBASE III Files | Michael C. Harris | The dBASE III fIle format is arguably the most enduring in the PC/MS-DOS world. Originated by Ashton-Tate in the 1980s, it has won widespread support among third party vendors of both enduser products and application development tools. The term "xBase" has been corned to describe third-party products supporting the dBASE fIle format. Today, specialized research databases are available in the dBASE .DBF format. They may contain many tens of thousands of records and occupy multiple megabytes of disk space. This paper discusses a technique for reading .DBF fIles into SAS data sets in the PC/MS-DOS and UNIX Sun-OS environments. The concepts may be applicable to o.ther operating systems as well | data step, variable, dataset, character variable, numeric variable, proc step | source | 7 |
| 244 | Using the SAS System for Error Control in Clinical Trials Data'bas'es | William C. Murphy, Howard M. Proskin, Gary B. Freeman |
In clinical trials. information is often collected via hard copy on case report forms, and must be entered into computerized databases prior to processing. The need to minimize errors in this transfer process is of the utmost importance ~ valid conclusions are to be obtained from the study. Consequently, a protocol for data entry and error detection must be established and strictly adhered to. The SAS DATA STEP can readily scan for illegal values as raw data is input. Duplicate data entry followed by the use of PROC COMPARE provides a way to locate discrepancies resulting from data-entry errors. Furthermore, PROC COMPARE can assist in providing a quantitative assessment of the number of questionable entries in the final data set. By using the SAS Macro language, ~ is possible to automate this process to accommodate almost any study protocol |
proc compare, proc step, macro variable, raw data, macro loop, macro definition | source | 3 |
| 245 | Using the SAS System in the Pharmaceutical Industry: Project Setup and Standardization for the Analysis of Clinical Trial Data. | Cam Huffman, Carl Arneson, John Horton, John Henry King |
At Burroughs Wellcome Co., statistical programmers in,the Clinical Statistics Department generally work on many different projects. Each project requires that different types of data be collected, managed, analyzed and presented. Usually, each project team has its own strategy of dealing with these responsibilities. In the past, this has resulted in programming inconsistencies and duplication of effort among SAS programmers. To overcome this problem, we have implemented a system to standardize SAS programming strategy and report generation within and across projects. The system includes a set ofSAS programs and macros for keeping track of titles, pre-processing data to create "analysis data sets" that contain both raw and derived data structured in logical and useful ways, and in~iali2ing SAS programs for a particular project. This system has helped us acheive greater programming efficiency and consistency and at the same time, has helped us to improve the quality of our work | macro variable, dataset, txt, proc step, proc report, title statement, medical history, sas programming, call symput, libname statement | source | 5 |
| 246 | Using the SAS System to Integrate Managerial Economics Concepts of Optimization | Seid M. Zekavat | proc step, proc corr | source | 6 | |
| 247 | Using the SQL Procedure | Kirk Paul Lafler | Using the SQL Procedure Kirk Paul Lafler Software Intelligence Corporation The SQL procedure follows most of the guidelines established by the American National Standards Institute (ANSI). In its current form, PROC SQL provides many features available in the DATA STEP and the MEANS, PRINT, and SORT procedures. An advantage of using PROC SQL is that it can often result in fewer and shorter statements than using existing DATA STEP and procedure methods. This hands-on workshop illustrates numerous examples of how PROC SQL and its many statements can be used. In particular, participants will learn how to create and modify tables and views, retrieve data using the SELEcr statement, and perform efficient queries against SAS System data sets. Brief History of Structured Query Language (SQL) Structured Query Language (SQL) has an interesting history: It is a universal language that was developed to easily access data that is stored in relational databases or tables |
proc sql, proc step, group by, order by, create table, relational database, select, update statement, where statement | source | 10 |
| 248 | Visualization of N-Dimensional Databases Using Parallel Coordinates | John J. Helly, Arthur L. Carpenter |
Conventional methods of exploratory data analysis em ploy tools ranging from two and three dimensional scatter plots and descriptive statistical summaries to multivariate | scatter plot, exploratory data analysis | source | 3 |
| 249 | Visualization of Porous Media | Leon L. Fedenczuk, Brian Wyvill |
We present a system for the visualization of porous media, which is based on a spatial subdivision technique. The media is represented as an interconnected set of cubic nodes that are organized in a structured fashion. The model .can represent ,the porous media at any scale, by introducing subdivisions as required within the limits of the hardware. A user can visualize an image of the 3D model by rendering nodes with different colors, which represent different parts of the media, or even as transparent or semi transparent nodes. In addition to viewing the models in three dimensions, the system allows a user to view arbitrary cross sections through the model. The developed system will allow users to monitor gas and fluid flow through models,· under various physical conditions. This can be achieved by applying a set of algorithms to find spatial connectivity between regions of the model | cross-sectional data, network optimization | source | 8 |
| 250 | Warning: Apparent Macro Invocation Not ResolvedTechniques for Debugging Macro Code | Veronica R. Walgamotte, Jeffrey Phillips, Derek Drummond |
macro variable, proc print, macro put, macro facility, macro quoting function, proc step, call symput, %if-%then-%else, error message, sas log, macro loop, macro definition, error, &variable | source | 6 | |
| 251 | Who Should Teach, What to Teach, How to Teach: An Illustration Using the SAS Macro Facility as an Example | Erika Mares Stone | This paper addresses two questions: first. tre general question of how to provide effective instruction in advanced SAS programming to a given group of SAS users and, second, the specific question of how to introduce the complex subject of the-SAS macro facility. The proposed solution to the first question is to encourage experienced programmers, working in the same environ· ment as the intended audience. to develop and teach classes. The benefits of teachers who are programmers belonging to the same group they are teaching are many, and go beyond the savings in costs associated with provid ing outside instructors. The second question is addressed by presenting excerpts from a programmer-designed introduction to the macro facility. The setting is a university research environment; the audience consists of students and programmers who use SAS to process and analyze survey data | macro facility, macro definition, dataset, input statement, proc step, proc freq, macro loop, libname statement | source | 6 |
| 252 | Why Forecasting Is a Waste of Time: The Three Aphorisms | Michael Gilliland, John LaBella |
The standard approach to improving forecast accu racy is to invest in more data, ~igger computers, fancier forecasting models, and more people to provide input. Yet for all the time and money spent in this endeavor, improvements have been scarce no magic formula has been found. We argue that the standard approach is fundamen tally misguided. It is often a mistake to invest heav ily in sophisticated forecasting systems because many processes cannot be forecasted well. Instead, effort should be focused on simplifying and smooth ing the processes being forecasted. By 'making the data forecastable' you will achieve belter forecasts with much less cost and effort. Using an example from the consumer products busi ness, we suggest a way to lower the volatility (and unpredictability) of sales pattems | cv, standard deviation | source | 6 |
| 253 | Why We Replaced DB2 Software with SAS Software as a Relational DBMS for a 30-Gigabyte User Information System | Stephen C. Scott | Every company has a need for a user information system where data can be presented for analysis. This system was originally designed as a normalized relational DB2 system to contain data from various databases. The users for this system are 100 analysts from departments throughout the corporation. However, in tests comparing access and CPU time, SAS software so outperformed the DB2 tables that the company elected to change from a relational database management system (DBMS) using DB2 software to one that uses SAS software. When tested against DB2 in accessing millions of rows of data, SAS software provided an 80 percent reduction in CPU time. This presentation describes the test plan and results of the comparison between DB2 software and SAS software as a relational DBMS. In addition, this presentation shows how we designed a user model DBMS with SAS software that contains one percent of production and, thus, allows quick testing of complex SAS programs | cpu time, dataset, proc step, relational database, sample | source | 10 |
| 254 | Writing SAS Code Behind The Scenes: A Conversational Macro Example | Deb Cassidy | Conversational macros provide a way for a non-SAS user to "write" SAS code when only base SAS is available. This paper will give you some advantages/disadvantages of this | macro put, macro variable, put statement, if-then-else, %if-%then-%else, macro loop, sas base, macro definition, proc step | source | 6 |
| 255 | Yes, the SAS System Can Read the Data | Barbara B. Okerson | One of the advantages of the SAS System is the ability to work with and process data from almost any format. This paper provides a discussion of how SAS processes data, followed bY,six examples that use SAS DATA STEP statements to create SAS data sets from six very different raw data files:. These examples are all real problems presented by users at Memphis State during the past year. All code used in the examples is included in this paper |
dataset, proc print, data step, proc step, do loop, mod | source | 5 |
This site serves as a convenient hosting location and does not work as a publisher. All copyrights belong to their respective owners.
The information on this site is provided for general reference purposes only. Whilst every effort is made to ensure that the site is up to date and accurate, no warrant, nor any responsibility or liability for, the accuracy or completeness of the content or for any loss which may arise from reliance on information contained in this site is provided. Links to other sites are provided for your convenience but no responsibility or liability for the content of those sites or of any external site which links to this site is accepted. References to any specific commercial product, process, service, or corporation are also for information only, and do not constitute endorsement, recommendation or favoring.
SAS is a registered trademark of SAS Institute Inc. SAS, Statistical Analysis System, and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc. in the USA and other countries. ® indicates USA registration.