HOME
QUICK SEARCH
AUTHORS
KEYWORDS
HELP
BUGSSUGI 1992 Papers - Index
263 papers. Source: SAS Users Group International conference proceedings.
| # | Title× | Authors× | Abstract× | Keywords× | Extras | Pages |
|---|---|---|---|---|---|---|
| 1 | "You Need This Done by When?"- Increasing Your Efficiency with the SAS System | Mark S. DeHaan | data step, proc step, put statement, proc printto, data statement | source | 4 | |
| 2 | A Binary Search for Non-Unique Keys | Derek Drummond | This paper will describe a binary search algorithm that allows users to specrty a range of key values. Most common binary search algorithms requirea single key value and will search for only one observation. In fact, these algorithms assume that each key value in the file to be searched can be found on only one observation (i.e., key values are unique). For our modified binary search algorithm, the key values do not have to be unique. The algorithm finds the first and last observations of a group of observations having the same key value. Also, the algorithm can search for a range of key values, finding the first and last observations that satisfy the range requested. Even if the file to be searched contains only unique key values, this modified algorithm can be used. This paper will be divided into three major sections | unique key, set statement, point= option | source | 5 |
| 3 | A Bootstrap Evaluation of the EM Algorithm for Censored Survival Data | George W. Pasdirtz | bootstrap, sample, proc step, mse, proc lifereg, likelihood function, rmse, response variable, regression analysis, normal distribution, normality test, proc calis, standard deviation, label statement, maximum likelihood estimation, macro definition | source | 6 | |
| 4 | A Comparison of the SQL Procedure with Conventional DATA and Procedure Step Methods | Kirk Paul Lafler, Fran Larsen |
If you have used the SAS System for any period of time, you are probably familiar with conventional DATA and PROC STEP processing to extract, merge, analyze, and present data. But you may not be familiar with how the SQL procedure can be used to perform these very same tasks. In fact, using the SQL procedure can often reduce the amount of code needed to perform a task as well as minimize the level of resources required by the system to process a request. As a method of comparing the SQL procedure with conventional DATA and PROC STEP methods, this paper presents application examples with their corresponding benchmark statistics to contrast the performance differences. The resulting differences are reported and performance strategies are recommended. 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, relational database, index creation | source | 5 |
| 5 | A Do-It-Yourself Guide to Creating a Data Entry Application Using SAS/FSP Software and Screen Control Language (SCL) | Veronica Fasullo, Felix G. Meale |
Having information collected on paper, then keyed to tape or disk in ASCII files, can be a waste of time, time which could be spent analyzing the data more thoroughly. Using SAS/FSP software, in our workshop at SUGI 17, we will show you how to create a working information collection system in about two hours. This system will provide on-line edit checks and display warning and error messages when data are entered incorrectly. SpecHically, you will learn how to: use PROC FSEDIT with the NEW and LABEL options; customize screens for data entry; set up initial, minimum, and maximum values; and wr~e a short SCL program to validate data entered |
proc step, dataset | source | 5 |
| 6 | A Guide to Modifying an Introductory SAS Course | Laura Ray-Wilson, Christine Stroup-Benham |
This paper covers both the explanation of the process involved in developing and teaching an introductory course to the SAS System for Personal Computers and the results of this restructured course. Historically, the course had been taught by instructors who were not regular SAS system users. Much emphasis was placed on teaching the SAS Display Man_ager thus reducing the amount of time available to teach the principles of the SA~ system. In the process of restructuring the course, certain questions had to be answered! 1)can the course be shortened; 2)~ho should teach the course; and 3)can the course be targeted to certain groups? The course was rewritten, with the OM being optional (as time permits), tested on 6 people from one office and taught by a SAS system user of 12 years | sas base, proc step | source | 6 |
| 7 | A Macro Procedure in SAS Software to Summarize Output of Regression Models | Chih-Ping Chou, Eric Yu I. Wang |
regression analysis, response variable, proc reg, proc calis | source | 4 | |
| 8 | A Menu-Driven SAS/QC Software Application | Mary B. Obenshain | label statement, catalog entry | source | 5 | |
| 9 | A Menu-Driven Table Generator | Joy Y. S. Jair, Rex C. T. Wu |
Instead of coding a SAS program each time when a demand is requested, a user firendly, menu-driven table generator is introduced. In the generator, users can define their tables by just selecting some condi tions and interested items he needed in the menu. The desired tables are produced to a great degree of flexibility according to the variate selections. SAS/ AF and SAS macro facilities incorporated with PROC TABULATE are fully utilized to meet the versatile requirements. A personnel system is used as an example and is under normal operation now. The results show great success. Preface SAS PROC TABULATE constructs table of descriptive statistics from class variables, analysis variables and keywords for statistics. It is not uncommon that the demands to produce different kinds of tables from the end users have become the major loads of an application programmer. Instead of coding program for each demand, a menu-driven table generator is needed for most of the cases |
class variable, filtering, proc tabulate, proc step | source | 7 |
| 10 | A PostScript Macro Library | Derek P. Morgan | cpu time, sas output, proc step | source | 6 | |
| 11 | A SAS Code Formatter | William S. Calvert, Malla R. Rao |
Most high level languages like C. Pascal. have code formatters which format source code for ease of understanding and debugging. SAS-. however. does not have such a tool at this time. We have implemented a simple code formatter to format SAS code. which recognizes commonly used constructs. We have used SAS for convenience in transporting the code across platforms, with no operating system specific statements. Our formatter illustrates text parsing and could be easily extended or modified to accommodate other standards | if-then-else, data step | source | 5 |
| 12 | A SAS Macro for Performing Analyses of Random Effect Models of Longitudinal Data Using the IML Procedure | Greg A. Grandits, Patricia Grambsch, Thomas Louis |
This paper presents a SAS" macro using the IML procedure to compute maximum likelihood (ML) and restricted maximum likelihood estimates (REML) and tests of hypotheses for random effect models of longitudinal data as described by Laird and Ware (I) using the EM algorithm (2). This model allows analysis of data with repeat observations on each subject (or unit) where each subject or unit is observed at several times and/or under different conditions. Unlike repeated measures analysis of variance this model allows for a varying number of observations per subject or missing data and for time-dependent covariates. Also, this model can be used for meta analysis adjusting for covariates where one is combining the results of several studies. There is currently no SAS" procedure to perform the analysis for this model. Input to the macro are the fixed (X matrix) and the random effect (Z matrix) factors located on a SAS" data set |
response variable, proc mixed, degrees of freedom, restricted maximum likelihood, covariance, likelihood function, repeated measures, repeated measures anova, proc iml, missing value | source | 6 |
| 13 | A SAS Macro to Extend the Friedman Test for the Unbalanced Case Plus a Multiple Comparisons Procedure | Kirk A. Easley | A SAS@ Macro utilizing the method of m rankings (or within block ran kings) generates the Benard-van Elteren test statistic which is an extension of the Friedman test that permits empty cells and any number of observations per cell. A multiple comparisons procedure for pairwise comparisons of treatments is included for use in conjunction with the Benard-van Elteren test Additional applications of this Macro include the Kruskal-Wallis test and the analysis of a two-treatment, two-period, four-sequence crossover model | dataset, missing value, friedman test, macro variable, anova, covariance, proc step, macro call | source | 4 |
| 14 | A SAS Program for Reproducing the Results from the ITEM Procedure | Shu-Yeng Wong | proc step, dataset, out=, standard deviation, correlation, point-biserial correlation, macro variable | source | 5 | |
| 15 | A SAS Program to Analyze Statistical Distribution of Airborne Material | D. Manco, F. Caiffa, T. Tirabassi, S. Corsi |
weibull distribution, proc step, cumulative frequency, proc sort | source | 5 | |
| 16 | A SAS Program to Compute Factors for One-Sided Tolerance Limits for a Normal Distribution | Susan Portugal | One sided tolerance limits for a normal distribution are one-sided (upper or lower) limits such that at least a proportion P of the normal population falls below an upper limit or above a lower limit with confidence y. When the parameters 1.1 and 0' of the nonnal distribution are both unknown, the limits are computed from the sample mean X (based on N observations) and sample standard deviation S (based on F d..§lgrees of free~m). Specifically, a one sided limit is of the form X + K S or X - K 8, where K is the tolerance factor that depends on r, P, N, and F. This paper presents a program for computing K based on an underlying noncentral t-distribution. The program uses the SA~ TINV function to compute an exact factor when the noncentrality parameter is less than 15.5. When the noncentrality parameter is 15.5 or more and SAS- TINV returns a missing value, the program calculates an approximate factor | normal distribution, sample, standard deviation, confidence level, missing value | source | 4 |
| 17 | A SAS Program to Create Output Files Used for a Randomization Code Distribution Procedure | Joseph M. O'Connor, Ute E. Schwiderski, Joseph J. Andary |
In order to do randomizations for drug studies at Bristol-Myers Squibb. a CMS REXX program was developed which builds and executes a series of SAS@ programs to produce the randomization output. Since this output needs to be packaged and distributed quickly and efficiently, a supplemental SAS program was developed to perform ~his task. This paper addresses the features of the SAS system that were used in the development of the packaging and distribution program. It runs in the SAS" 5.18 display manager and uses the SAS/FSP procedure FS~DIT as a data entry interface. Applications of the SAS macro language were found to be particularly useful in creating a relatively simple and flexible program |
randomization, macro variable, index, sample, proc step, call symput, macro definition | source | 5 |
| 18 | A SAS System: From Mainframe SAS Data to Spreadsheet via the SAS System for PCs | Samuel D. Calhoun | The use of spreadsheets to store and analyze data has been increasingly accepted in at the Economic Research Service. This has given the Agency a great deal of flexibility in processing and analyzing data more efficiently. A particular need for spreadsheets was brought about by the public's demand for world oilseed and products trade data. The data are stored on a mainframe system in various SAS" datasets. The data management system on this mainframe uses both EASYTRIEVE"1l1 and SAS(l1 to extract and combine different sources and attributes such as detailed import and export trade between countries from the United Nations Conference on Trade and Development (UNCTADI and the Food and Agriculture Organization of the United Nations (FAOI. When both partners (importer and exporter) report to UNCTAD on trade, the importer data is given preference to the exporter. The import preference data are then converted to ASCII data and downloaded to the PC | dataset, batch job | source | 4 |
| 19 | A SAS Window Application for Generating Follow-up Clinic Schedules and Monitoring Accrual in a VA Cooperative Study | Joan Derrico | In the Personal Computer environment an interactive data entry WINDOW system was designed using the SAS'" System. This application was written to facilitate the generation of schedules for patient follow-up clinic visits and graphically monitor patient accrual in two Veterans Administration Cooperative Studies Program Clinical Trials. The use of customized WINDOWS as data entry screens proved to be a quick and easy way to enter pertinent study data immediately into the system and to trigger the generation of patient accrual plots using SA.S/GRAPH'". The DATASETS procedure with the APPEND option aided in updating the databases. This paper discusses this system and displays the data entry screens and output created by it |
randomization, clinical trial, sample | source | 6 |
| 20 | A SAS/IML Program to Simulate Data with Nonnormal Distributions | Chih-Ping Chou, Shinn-Tzong Wu |
correlation, sample | source | 3 | |
| 21 | A SASS Interface to System 1032: Writing a User-Maintainable Interface to Data Management Systems | Allan D. Rosen | Many SAS users analyze data maintained with data management systems from independent vendors. Interfaces to access these data sets in SAS may be inadequate, temporarily incompatible, or non-exislenl. Writing a conversion interface in the SAS macro language is recommended when it is desirable for SAS users to maintain control over the interface. Criteria for this decision include the commitment of vendors and central computing services to maintain a compatible interface. Strategies for writing such macros are discussed. SiS, a family of macros, implements these strategies to provkJe a SAS interface to System 1032. a relational data base management system for a VAXNMS environment. SiS is used and maintained by statisticians, epidemiologists, and other SAS users on several projects at the Epidemiology Data Center at the Universily of Pittsburgh | dataset, variable, data step, data governance, macro facility, macro definition, proc step | source | 5 |
| 22 | A Simulation Study of Estimators for Rates of Changes in Longitudinal Studies with Attrition | Fong Wang, Mary Lange |
Many longitudinal studies and clinical trials are designed to estimate rates of change over time in one or more outcome variables in several groups. Most such studies have incomplete data because some patients drop out before completing the study. The missing data may induce bias and inefficiency in naive estimates of important parameters. We used Monte Carlo | likelihood function, missing value, missing completely at random, restricted maximum likelihood, sample, clinical trial, monte carlo, sample size, weighted mean, covariance | source | 6 |
| 23 | A Supplemental SAS Function for Computing Moving Average | Shi-Tao Yeh | time series analysis, moving average, sample, time series data, macro definition | source | 3 | |
| 24 | A Tool for Compiling DATA Steps in Large SAS Systems | Herbert C. Reynolds | The Stored Program Facility available with SAS Release 6.06 enables data steps to be compiled and stored. Compiling and storing data steps eliminates compiling the DATA STEP each time the program is executed therefore saving CPU time and cost. It is very easy to compile, store, and execute a few data steps in one or two programs, but doing this for many programs in large systems the process can become tedious and unwieldy. To facilitate the compilation of many data steps in hundreds of previously uncompiled programs in a large SAS application, a set of command language programs were developed to do all of the tedious work for the programmer. By following a few programming guidelines and using the command language programs, the task of compiling, storing, and managing stored programs in a large application becomes very easy |
data step, data statement, proc step, proc sort, proc print, cpu time, keep statement, rename statement | source | 6 |
| 25 | A User-Friendly Front End to Access GDGs: Automatic JCL | Matthew Casella | This paper describes a dynamic database creator used within a Decision Support Group. The database creator automates the labor intensive and time consuming efforts of collecting/providing voluminous amounts of historical data. It presents a user with a selection list, developed using SAS( AF soft ware, that contains generation datasets and their creation dates. A user can point and click on each dataset to automatically create JCL that will create the historical data. The JCL invokes a robotic device to mount each cartridge. This system runs in conjunction with Tape Management Software (TMS) which stores all cartridge information on file | proc step, proc sql | source | 6 |
| 26 | A User-Friendly SQL Query Generator Using SAS/AF Software and SCL | Bert Cave | proc sql, where statement, group by, order by, proc step | source | 6 | |
| 27 | Ad Hoc Access to Diverse Databases for the Computer Illiterate | Eric Larson | A marriage of SAS / AF and SQL to provide seamless ad hoc access to a collection of databases for the computer illiterate. This menu driven system developed uses SAS/ AF and SQL to enable the computer illiterate user to access a collection ofDB2 Tables and SAS datasets. Data from various datasets are joined, without user intervention, on the basis of control information contained in SAS datasets. The user selects the desired query dimensions and query variables. The system submits the sequence of SAS steps that are necessary to gather the requested data and present it in a report or graphic form. The user can sort the data, customize the output format, develop new variables as expressions using extracted data, and summarize the extracted data through additional menu selections. The paper describes the use of SAS / AF and SQL techniques that were used to develop this system under SAS 6.06. Subsequent improvements to the system, that were made possible under SAS 6.07, are discussed | proc step, proc sql, dataset, feature selection, proc print | source | 8 |
| 28 | Adding the SAS Patriot to Your COBOL Arsenal | Douglas A. Bell | Many SAS users coexist with COBOL users who don't reauze how much time and effort. and therefore expense, could be saved by using SAS. $AS has a ~bad rap~ of being only a statistical language. If $AS users coukj show COBOL users how easy SAS Is to learn and how useful n can be to them. the COBOL users could be more productive by using SAS for many applications currently being written in COBOL | _eof_, input statement | source | 2 |
| 29 | Advanced Programming Topics for the SAS/ACCESS Interface to DB2 Using the New Features of Release 6.07 | Bradley W. Klenz, Charles A. Jacobs |
proc step, dataset, modify statement, update statement, macro variable, by group processing, sysrc, select, sas log, libname statement, transaction data, do loop, group by, order by, key, where statement, flat file, proc sql, proc contents, rename statement, variable, date variable, macro facility, db2 access | source | ||
| 30 | An Advance in Geographic Information Systems Using SAS Software: Address Level Lat-Longing | Doug Surch | proc step, proc gmap | source | 6 | |
| 31 | An Alternative Convergence Criterion for the NLIN Procedure | William Potts | The convergence criteria used with the iterative nonlinear least squares estimation methods in the SAS NUN procedure can be satisfied when the final estimates are not the least squares estimates. In contrast, the relative offset convergence criterion uses geometric properties of the nonlinear least squares estimates to measure true convergence. A SAS macro is given which calculates the relative offset convergence criterion so that the results of the NUN procedure can be validated. Nonlinear Least Squares In nonlinear regression, a function, f(X1, ... ,~,8), is fitted to an n-vector of responses, Y. in order to estimate a p-vector of unknown parameters, O. The n vectors, Xi (i = 1, ... ,k), are the fixed regressors. The set of values of the regression function for all values of 8 in the parameter space defines a p-dimensional surface called the solution locus in the n-dimensional sample space | estimate, regression analysis, response variable | source | 4 |
| 32 | An Easy-to-Use Double-Entry Routine | Trent L. McDonald | The presence of entry errors in a data set can effect both the accuracy and precision of data. To combat errors made during the entry process, double-entry routines have been developed. A general, easy-ta-use, double-entry routine implemented inSAS" and SAS/FSP" software is presented. SAS/FSP is seen to provide an efficient interface for defining, entering, and updating SAS data sets. Enhancements to provide for data validation are discussed, as well as modifications when SASIFSP is not available. In one implementation of the routine, an average of 53 out of every 10,000 values did not agree between first and second entry |
proc step, macro variable, character variable, data step, numeric variable, %if-%then-%else | source | 6 |
| 33 | An Executive Information System Prototype at Long Island Jewish Medical Center | Lewis S. Goldson | The goal of any executive information system is to increase efficiency of operations in an organization by providing a tool that makes information available to enhance the decision making process of key employees and bring a unique perspective to critical issuesThis paper presents an | dataset, batch job, macro definition, macro variable, %if-%then-%else | source | 6 |
| 34 | An In-House SAS Users Group: Facilitating the Transition to Version 6 | Marcia S. Murto | proc step, sas base | source | 5 | |
| 35 | An Integrated Clinical Information System for Pharmaceutical Research | Martin J. Rosenberg | Clinical information systems (CIS) have traditionally concentrated on study definition, data entry, data security, and data retrieval; and their use has typically been the province of computer professionals, such as clinical data managers, biostatisticians, and programmers. Spurred on 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. These efforts have included systems for remote study monitoring, in-house medical review, and computer assisted NDA's and PLA's (CANDA's and CAPLA's) | crf, clinical trial, audit trail, data validation rules, proc means, clinical data, key, data review, data cleaning | source | 7 |
| 36 | An Interactive SAS/AF System for Assessing Bioequivalence | Deborah Dukovic | Testing the equivalence of treatments in a two-period crossover design setting occurs often in the pharmaceutical industry. These studies usually compare the rate and extent of absorption of two different drugs or two formulations of the same drug, where one is the accepted standard therapy. The commonly accepted rule for concluding bioequivalence requires the 90% confidence interval for the mean tesVstandard ratio of AUG and Cmax to be contained in the range (80,120)%. Sample size is usually determined based on meeting this criterion. In order to improve efficiency and consistency in the analysis and reporting of data from bioequivalence trials, a SAS/ A~ application system for analyzing data from a two-period, two-treatment crossover design was created. The system contains four main sections: sample size/power, analysis, bioequivalence and utility. With the expanded capabilities of SASIAF in Version 6 and the use of SCL, the system is very powerful, yet extremely user friendly | confidence interval, sample, sample size, anova, dataset, crossover study, cv, proc means | source | 6 |
| 37 | Analysis of Incomplete Categorical Repeated Measures | Charles S. Davis | The SAS@ CATMOD procedure is a powerful and general tool for the analysis of categorical data. Many widely-known and routinely-used applications are described in the SAS/STAT U.er'8 Guide and SUGI Proceedings. However, the use ofPROC CATMOD in analyzing repeated measures categorical data is less well known. In particular, existing documentation has fo cused on the somewhat unrealistic situation in which the data are complete. In practice, the occurrence of missing data is quite conunon in repeated measures studies. This paper describes and demonstrates the use of PROC CATMOD in analyzing incom plete categorical repeated measures data. An extension of the methodology for testing whether or not subjects with incom plete data are adequately characterized-by the same model as subjects with complete data is also discussed |
repeated measures, missing value, proc step, sample, estimate, longitudinal data, data science, response variable, proc freq | source | 6 |
| 38 | Analytic Modeling in SAS/INSIGHT Software | Heman Robinson | Analytic modeling software has evolved from line-mode, batch-oriented systems to graphical, direct-manipulation interfaces. Mouse- and menu-based input eliminates complex command syntax and makes users of these new interfaces highly efficient. SAS/INSIGHT 1M software provides an efficient environment for exploratory data analysis and model building. This paper describes exploration and modeling capabilities in current and future releases of SAS/INSIGHT software |
scatter plot, ancova, proc glm, predictor variable, histogram, covariance, regression analysis | source | 6 |
| 39 | Analyzing Business and Economics Data with SAS/ETS Software | Mark Little | This paper discusses analyzing business and economic data with SAS/ETS software. The paper surveys some of the different kinds of data analysis problems found with business and economic data, and the need for special data analysis methods for these kinds of problems. The paper discusses the tools SAS/ETS software pro vides to meet these needs |
time series analysis, regression analysis, proc step, time series data, proc arima, response variable, data science, predictor variable, seasonal adjustment, seasonality, likelihood function, proc forecast, proc freq, correlation, proc calis, output delivery system, proc esm, proc reg | source | 9 |
| 40 | Analyzing Multiple Response Questions: Facilitating an Existing Method and Presenting an Alternative | Paul E. Pope, Darrell Fannin |
This paper addresses the difficulties associated with analyzing multiple response data with SAS' software. After an examination of the existing | proc freq, if-then-else, proc step | source | 6 |
| 41 | Appeal to a HIER Authority: Graphing CA-7 Scheduling Dependencies | Brent Turner | The Computer Associates' (CA) scheduling product, CA-7, provides a narrative report of antecedentl consequent job dependencies, but in a somewhat congested presentation. The CA-7 output report is used as the input rue to a SASprogram which produces two SAS data sets from the report text. The observations in the ftrst data set each contain the job name, its antecedent job name, and, if there are dependencies, a generated reference number. Those in the second set contain the job name, with its dependencies (and the same respective reference number) if any. By using multiple SET statements in a DATA STEP, each observation in the ftrst data set is compared with every other observation in the same data set to determine job sequencing. The SAS supplemental procedure HIER produces a hierarchical diagram showing the jobs' schedule, where dependent job names are concatenated with their respective reference numbers |
set statement, conditional do loop, data step | source | 6 |
| 42 | Are We Having Fun Yet? Human Factors in Designing a SAS/AF Job Submission System | Jeffrey Phillips, Todd Peer |
batch job, libname statement | source | 9 | |
| 43 | Are Your Manufacturing and Distribution Systems Optimal? Exploiting Networks to Solve Linear Programming Models Faster | Maria Lupetin, Trevor D. Kearney |
This paper gives a brief description of linear programming (LP) and its applications. It discusses how to look for an imbedded network model with side constraints within an LP model. Models formulated as network models are solved several to hundreds times faster and with less memory than models using the traditional LP solution methods. The capabilities and enhancements of the SAS/OR NETFLOW PROCEDURE NETWORK solver are discussed |
proc step, linear programming, network optimization | source | 5 |
| 44 | Artificial Intelligence and the SAS System: Why You Have to Teach the SAS System about SEX! | David B. Malkovsky | This paper presents a very brief survey of the areas that comprise artificial intelligence (AI) research (knowledge representation, natu ral language understanding. expert systems, and so on). The majority of the paper describes AI researCh done at SAS Insti tute and a new product, SAS/ENGLISW software, which is a result of that research effort. SAS/ENGLISH software provides users with a means of asking for information about their data by typing English questions using a user-friendly interface that avoids a computer specific language or syntax. The importance of knowledge acquisi tion to make this interface work is discussed. The subtitle of the paper is used to draw attention to the need for teaching AI programs the words and concepts that humans take for granted when they query each other in normal conversation |
artificial intelligence, dataset | source | 6 |
| 45 | Artistic/Aesthetics | Joris Cauquill | proc gmap, proc gplot | source | 4 | |
| 46 | Assessment of Interrater Agreement for Multiple Nominal Responses among Several Raters | Chul W. Ahn | An interrater agreement coefficient is computed using a proportional overlap | confidence interval, proc calis, proc freq | source | 6 |
| 47 | AUTO WRAP/TABS: Are These Parameters a Limitation on IBM Terminals? | Marie E. Tennant | Ono of tho most useful features of Word Processing software packages is the ability to continue to type without thinking about margins or newline decisions. It was originally reported in Usage Note Y6-FSLETIER-Ot 11 that the AUTOWRAP and TABS parameters, when used with the FSLETTER procedure ot SAS! FSp, were ignored on IBM terminals. Not wanting to believe that IBM had not addressed this issue, I pursued the subject with the IBM technicians assigned t~ Pennsylvania Higher Education Assistance Agency (PHEAA). As it turns out, 18M'S Entf}' Assist provides improvements for display operators who enter and edit text material. Although, it is still true that the SAS parameters, AUTOWRAP and TABS, are ignored, IBM's Entry Assist provides the same functions |
source | 5 | |
| 48 | Automated Workload. Forecasting - A Large Complex SAS Application | Lance Pedriana, Fay Wong |
In the area of computer performance and capacity planning, large amounts of operational data often have to be analyzed. Data analysis using SAS has been documented many times over the years. However, the most critical area of capacity planning-forecasting resource requirements--has received minimal attention as published papers or even as an area of development by most installations. Forecasting has usually been little more than a simple trend analysis. Workload forecast data is not readily available in the way IBM's SMF (System Monitoring Facility) or RMF (Resource Monitoring Facility) is. A considerable amount of an analysfs time is needed to solicit. organize, and analyze forecasting data. In installations that process a large number of unrelated applications, this process is very time-consuming and tedious. This paper presents a SAS system that has been developed to automate the workload forecasting process as much as possible | capacity planning, reference data, seasonal adjustment | source | 8 |
| 49 | Automatic Documentation of COBOL Systems | John D. Smith | sample, input statement, label statement | source | 6 | |
| 50 | Automatic Documentation: Using SAS Software to Create the Data Dictionary, Data Structure, and Process Flow Tables | Arthur L. Carpenter | The documentation of SAS programs and data sets is critical, yet it is always the last thing that is done (when it is done at all). Documentation that does exist is rarely updated- and often not accurate. Usually this is because the process of producing and maintaining the documentation is too labor intensive and too time-consuming. If we expect our programmers to produce and maintain the documentation for the programs they wr~te, the documentation process needs to be as automatic and painless as possible. WellQocumented programs, and especially systems of programs, should include a Data Dictionary (definition and attributes of each SAS variable and an exhaustive list of the data sets in which it occurs), a Data structure Table showing the variables contained in each SAS data set and a Process Flow Description Table. This paper discusses the types of information expected in well-structured programs and their associated documentation | data dictionary, sample, proc contents, proc step, dataset | source | 6 |
| 51 | Best Presentation of Data-Color | Perry Watts | proc step, proc format | source | 9 | |
| 52 | Best Presentation of Data-Monochrome | Richard L. Berg | scatter plot, box plot, %macro parameters | source | 7 | |
| 53 | Beyond Executive Information Systems | Neil H. Howard, Richard W. La Valley |
data wrangling, management reporting | source | 7 | |
| 54 | Blocked Randomized Assignment of Treatments to Patients in Clinical Trials | Maria A. Deloria, George F. Reed |
Randomization is an essential device for eliminating bias when assigning treatments to patients in clinical trials. It is often carried out in blocks of consecutively accrued patients to ensure short-term balance in numbers of patients in each treatment group. A short, easy and adaptable method is presented for producing blocked randomized treatment assignments for patients in clinical trials. This method, based on PRoe PLAN, is a manageable alternative to the lengthy algorithm of SASS code otherwise needed to perform the same task. Examples us ing PROe PLAN to create treatment assignments with varying block sizes and with unequal treatment allocation illustrate the flexibility of the method. The PUT statement and FORMAT procedure are used to produce forms containing patient ID numbers and their respective assigned treatments, which are suitable for direct use by study staff |
randomization, proc step, proc format, clinical trial | source | 6 |
| 55 | Bootstrap Estimation of the Variance of Percentile Regression Parameters | Michael L. Smith, George Howard, Gregory W. Evans |
Just as percentiles provide an extended description of univariate data, percentile regression may provide a superior description of bivariate data by establishing the relationship of a covariate x to the percentiles of a response variable y. Recent literature on percentile regression has discussed extensions to situations with multiple covariates, but has not addressed the problem of obtaining variance estimates for the regression parameters. A variance/covariance matrix for regression parameters based on a specific percentile (P10/J.(Y) = f(x) for example) can be obtained by bootstrapping the original data. Using SAS/LML' software, a method has been developed for finding the bootstrap estimate of the variance for percentile regression problems based on a polynomial regression equation. A complete description and source code for this method is provided, along with an example application |
regression analysis, bootstrap, covariance, proc reg, sample, response variable | source | 9 |
| 56 | Building Data-Driven Applications Using the SAS Applications System: Selected Techniques | 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 describes a system architecture by which an application can be maintained and enhanced through changes to data and parameter files. An application designed this way allows for more flexibility and lower maintenance than one in which code is written to perform specific functions. The generalized code in a data-driven system performs under the direction of parameters, allowing any changes to be implemented by altering values in parameter files instead of changing the source code. Examples include such cases as adding new reports changing or reordering menu items * controlling user access and security. Sample programs that make use of SAS/AP and SAS/FSP" soft ware using Screen Control Language(SCl), the Sal procedure, and the SAS macro facility are presented and described in detail |
sample, proc step, proc sql, proc format, macro variable, macro facility, macro definition | source | 7 |
| 57 | Calculating an Exact Permutation Trend Test Statistic Using the SAS System | Steven Bailey | The Exact Permutation Trend Test is an exact lesl for ordereg alternatives that is not currenHy available in Ihe SAS fR system. This paper reviews the use 01 the Exact Permutation Trend Test and presents an algorithm thai can be used 10 calculate the p-value. Algorithm efficiency and extensions are also discussed | macro variable, macro conditional | source | 5 |
| 58 | CDMSTAT. A Case Report Forms Tracking System using SAS/AF and SAS/FSP Software | Faith Reneé Sloan, H. Jean Lieverman |
1:n the pharaaceutica.l research industry, .it. .is asaantial to .0Ditor each study site for all ongoing drug studies .a to the nuaber o~ ca report forma(CRFa) received, PROC __ d. and/or qu_tioned. With this information readily available, relevant parti_ a~ aware of sita& that ex-Il adharinq to protocol. with rel.ative ea or site. that ae ** to be ·problawatic·. Que.tiona can be .aked early and proble_ may be resolved in a t1111s1y raahion. The cl..1nic.l. Data Kanage_nt Status(CDNSTAT) By.tea was das1gned and daveloped as a tool. to _at the needs of the Cl.lnlcal Data Kanaqa_nt Department at Parka-Davia to track the status of al.l CliFa received 1n howae |
proc step, phase iv | source | 3 |
| 59 | Checking Out-of-Range Data | Aileen L. Yam | outlier, proc step, proc univariate, proc freq, proc print | source | 4 | |
| 60 | Computer Aided Name Matching | James Sidlo | Name matching hospital names is a programming challenge since Hospital names are more complex than one's own name. Hospital names tend to be written in any number of ways with name qualifiers added in or left out. word transpositions or misspellings, and arbitrary abbreviations, all the result of the writers whim. This is apparent in the examples given below. Personal Name Joan Mary Ford J.M. Ford J. Ford Hospital Name Baptist Hosp~a1 East Bapt Med Center East Baptist East Hospital Matching. whether it is done manually or electronically. requires keywords (high lighted in the examples above) to search by. The more unique a keyword. the greater the possibility of finding a single match. This most basic goal of computer aided name matching has been enhanced by an integrated solution presented in this paper. This was accomplished by combining human and computer resources using the tools supplied by Base SAS 6.06 | sas base, if-then-else | source | 5 |
| 61 | Control and Monitor Facilities for Multi-User Access | Randy Horne | access control, audit trail, sysrc | source | 5 | |
| 62 | Control of Batch Program Execution with the SAS System | Tyler Witko, Thomas R. Hoffman |
Information systems consisting of logically related data sets updated by batch programs submitted independently by multiple users require a method of controlling batch job execution. Control is required to ensure updates to the data sets in proper order. By using the SAS System to write batch job execution requests to a sequential file and then processing the file with SAS Macro Language, a control module was developed that effects batch job scheduling, sequencing, and documenting of job execution histories for analysis. Although this module was developed for a pharmaceutical application on an IBM mainframe running under MVS, ~ could function for any type of information system on any platform that permits batch jobs to subm~ other batch jobs for execution | batch job, clinical trial | source | 5 |
| 63 | Controlling which files are displayed in the PROC ACCESS window | Bruce F. Gilsen | proc step, sas library | source | 2 | |
| 64 | Corporate Data: Bringing Together the Islands of Information with SAS/ACCESS Software | Emily P. Wallace | Since many companies have more than one database management system, an important concern is how to bring together the data from these islands of information in the most efficient and effective man ner. SAS/ACCESS software provides a transparent link to your database software. These interfaces allow everyone in the company from end users to the database administrators to use the tools that suit their activities to reach the database data and to perfonn their analysis, reporting, or querying functions. This paper examines the ways that SAS/ACCESS software integrates into SAS" applications to deliver database information transparently |
dataset, proc step | source | 4 |
| 65 | CPU and DASD Impacts of Release 6.06 of the SAS System | Michael J. Schilling | = SAS@ 6.06 introduced options for tuning SAS@. Some options are for tailoring SAS@ for the environment it will be executing and some options for influencing performance during execution. This paper summarizes the results of the benchmarks and tuning during the installation of SAS@ 6.06 at Burlington Northern Railroad. The benchmarks were done in a host MVSjXA environment for both MBatch | cpu time, proc step, proc copy, bufsize, raw data | source | 7 |
| 66 | Creating a Data Dictionary for Large SAS Data Sets | Steven A. Wilson | In many processing environments, SAS programs are commonly used to read data from a flat file and "load" it into a SAS data set for analysis purposes. The users of this loaded data are often provided with scant information about the data set, though. Although PROC CONTENTS provides descriptive information of the data set, this information is generally not enough to satisfy the user's very real concerns about how and where the variables in the data set were obtained. In Version 6 of SAS, this problem is intensified because Version 6 SAS data sets do not contain the source code used to create them. This paper will discuss methods that can be used to automatically construct a comprehensive on-line data dictionary for the "loaded" SAS data set. The resulting data dictionary can be viewed using |
dataset, data dictionary, missing value, flat file, savings account, proc step | source | 6 |
| 67 | Creating a Developer's Toolkit: Facilitating the Development of SAS/AF Applications | Daniel Gronell | This paper analyzes several types of SAS! AF application development tools, Its intent is to provide some insight into tool kit development and use, and to provide readers with some useful tools for their own tool kit, It is suited to developers having moderate experience with the SAS! AF product, Screen Control Language (SCL), and the SAS" software macro facility, | sysrc, macro definition, macro facility, %if-%then-%else | source | 6 |
| 68 | Creating Comma-Delimited Files from SAS Data Sets for Downloading to Microcomputers | Karen C. Patterson | call symput, delimited file, macro definition, proc contents | source | 3 | |
| 69 | CUMULUS, Graphic Representation of Factorial Planes | Danielle Grange, Michel Ringenbach |
SAS and SPADN are the main software used for factor analysis at the CNRS Computer Center in Strasbourg. But neither of them produce printouts of factorial planes that are suitable for publishing. Moreover the information needed to interpret the results is not given on the plots, but scattered through different tables. Cumulus enables the user to obtain a high resolution graph of factorial planes. Active and supplementary variables (or individuals) can be drawn as well as the attributes of the plane (Le. eigenvalues, ratio of inertia). The cloud of individuals can be split into subgroups according to the various categories of a variable or to the results of a cluster analysis. Cumulus permi ts : to obtain clear labels of up to 20 characters to separate overlapping labels to choose symbols of points to work out all the graphic elements, titles, footnotes, axes, inertia ... * to have full control of fonts, character height and colors | proc corresp, proc step, proc gplot, correlation | source | 7 |
| 70 | Dances with Data | Ginger Barlow, Angela S. Ringelberg |
Under the confines of different line sizes and page sizes, how do you produce custom reports that are 'reader·friendly"? What do you do when long character strings wrap to the next line without taking into account the structure of the string. (Le. spacing. punctuation, etc.)? How do you force pages to break between groups when all the observations in the next group will not fit on the current page? If numeric data is stored as character, how can you line up the decimal places? In the process of customizing reports, We have developed several tools and techniques to manipulate data using the DATA _NULL_ step |
proc print, proc step, data step | source | 6 |
| 71 | Data Exchange Between the SAS System and Microsoft Excel | Louise Weiler, Raymond Wan |
excel, variable, macro definition, filename statement, missing value | source | 6 | |
| 72 | Data Set Manipulation with Screen Control Language (SCL) | Marge Scerbo | where statement, sample, dataset, sysrc | source | 8 | |
| 73 | DATA Step Performance | Pat M. Herbert, Jeffrey A. Polzin |
For production sites, several changes in Release 6.07 will help reduce DATA STEP CPU usage. One of the first concerns in produc ing Release 6.07 was SAS data set 1/0 because the vast majority of OAT A steps produce SAS~ data sets and a large number of them also read SAS data sets. Another major concern was INFILE perfor mance. From tests on Release 6.06, it was clear that Release 6.06 unformatted external file input was faster than Release 5.18 exter nal input. This lead to benchmarking of aU informats, which were one source of poor performance. Another source of poor INFILE performance was column positioning. In Release 6.07, DATA STEP column positioning has been improved, especially positioning expressions such as offset+5 , +5 , and @5. Similarly, informats have been improved by up to 700%. Special attention has also been paid to OAT A step functions |
cpu time, dataset, missing value, input statement | source | 4 |
| 74 | Data-Checking Methods in a Data Entry System Based on SAS Software: Exploring the Possibilities | Sandi Tennyson | The SAS system offers a flexible applications development environment with a wealth of choices for the developer. With reference to our recent experience in developing a data entry system using SASIFSP, various data checking methods, including the SAS/FSP Special Attributes section. the use of SAS Informats. and special elements of the Screen Control Language (SCL) will be discussed in terms of bow Ihey can be used to address typical data entry systems development requirements. The data checking methods are analyzed in terms of how they handle 1) single field data checks, 2) cross-field data cbecks, 3) missing data. 4) the decoding or expanding of data entries, and 5) error messages, alarms, and the overriding of errors. Designing flexible systems will also be addressed, and discussion of the costs in tenns of systems developers time is included |
proc format, data cleaning, macro variable, proc step, error message, sysdate9, missing value, macro definition | source | 5 |
| 75 | Database Features Extend the Scope of SAS/SHARE Software | William D. Clifford | The role of SAS/SHARE software in the SAS System is being expanded to include major database features useful in both single- and multi user applications. Such features include integrity constraints (the specification of data validation criteria for all data inserted/added to a data set) and an audit trail (the ability to keep track of which users updated what observations and when they did it). The audit trail can optionally be used to reconstruct a data set from an archive copy |
audit trail, constraint, key, dataset, proc sql, data cleaning, create table, foreign key, where statement, missing value, proc step | source | 6 |
| 76 | DateDiff - More Than a SAS Macro | Jerry Kagan | sysdate9, macro definition, %if-%then-%else | source | 6 | |
| 77 | Departmental Automation Using Release 6.06 of the SAS System | Michael Sternberg, Peter Yribar |
Throughout the automotive direct marketing industry there is increasing pressure to produce high quality research under very tight time and oost oonstraints. Out of these ooncems arose the need for an interactive computer system which could perfonn the myriad tasks associated with automotive research yet still meet strict time and cost constraints. The system developed is known as ARAS (Automotive Research Analysis System) and it addresses the above ooncems by providing an online envirorunent which automates many of the routine functions associated with automotive research | response variable, macro variable, temporary dataset, dataset, sample, data cleaning | source | 6 |
| 78 | Design Features of a Clinical Information System | Bernadette H. Johnson | At Pharmaceutical Product Development (PPD), we needed a flexible system to facilitate data management and biostatistics activities. The task was to develop a common interface to access multiple and fluctuating numbers of clients and studies per client. PPD Clinical Information System (PPDCIS) is developing as a data driven application that uses a central database to: control access to data across clients and studies and adjust immediately to an addition or deletion of a client or study. This paper will give a general overview of the application features of PPDCIS and describe the design and advantages of the application's central database, PPD Project Access and Security System (PPD-PASS) | adverse events, data governance | source | 6 |
| 79 | Desktop Strategies: Should I Go to OS/2 Windows, or UNIX? | John Sims, David C. Brumitt |
source | 2 | ||
| 80 | Determination of Response Surface Function Using Repeated Measures | Denise M. Moyse, David Henry, Hong-Jiong Shi, Robert S. Boger, Clement Maurath |
anova, dose, clinical trial, repeated measures, change from baseline, response variable | source | 6 | |
| 81 | Developing Relational Database Interface Applications Using Version 6 SAS/FSP Software | Suzanne D. McCoy, Elizabeth F. Henrickson, Bill Lawson |
proc step, proc sql, data wrangling, relational database | source | 10 | |
| 82 | Developing SAS Applications Using Information Engineering Methodology | Priscilla E. Van Grevenhof, Roger Mueller |
The Mayo Clinic adopted IE (Infonnation Engineering) methodology as the standard development methodology for systems in 1988. The Texas Instrument IEF (Information Engineering Facilityr Toolset was the automated toolset chosen for application development. For institutional data the target platfonn is DB2.~ In the Department of Health Science Research IE methodology is also used but the target platfonn may not be DB2. The many research projects done at Mayo are short lived, from I to 3 years, and one of the tools used is SAS." SAS as the target platform is chosen because of its flexibility and short development time. These attributes match the needs of the research community. This poster will show how IE methodology is used to develop a SAS application and how IEF objects are related to SAS objects | libname statement, dataset | source | 3 |
| 83 | Development of a User-Friendly Mapping System | Peter J. Lund | libname statement, dataset, macro variable, proc step, temporary dataset | source | 8 | |
| 84 | Development of the RELIABILITY Procedure Using SAS/TOOLKIT "Software | Shobha Choudhari, Chintamani Sahoo |
The paper describes the implementation of the RELIABILITY procedure using SAS{TOOLKlT1M as a programming tool interiacing with FORTRAN language on V IV<NMS 1M platform; This application software provides a user written procedure, PROC REL, in the SAS 1M software base to conduct the reliability tests for various items in a given scale. The procedure is portable to other platforms supported by SAS{TOOLKIT software and by FORTRAN-77 language without major modifications in the source code. The REL procedure reads a SAS data set, periorms reliability analysis for the given items, and prints the |
correlation, anova, proc step, proc means, sample | source | 6 |
| 85 | Diagnosing Error Messages in Version 6 of the SAS System: But What Does It Really Mean? | Sally J. Painter | The SAS Guide to Problem Solving and Error Messages, Version 5 Edition has proven to be an excellent source of information for users as they diagnose problems with SAS code. This manum allows you to find an error and then follow the guidelines given for correcting the error in your program. However, with the introduction of Version 6, a new error numbering scheme was born. Instead of list ing a single error number, you may see more than one listed together on your SAS log. At first glance, it may seem more difficult to diagnose your own error messages. This paper addresses error-message handling in Version 6 of the SAS System. The difference between compile time and execution time errors will be discussed, as well as how users can utilize this information to diagnose their own problems. In addition, you will learn how to take the new information given in a Version 6 error message and use it to your benefit as you correct programming problems in your own programs or those of your users | sysrc, put statement, sas log, error message, automatic macro variable, abort, macro variable, proc step, return | source | 6 |
| 86 | Diagnostics for Model Selection in Nonlinear Regression Based on the Concept of Closeness to Linearity Using SAS/IML Software | Suzette Blanchard, Martin S. Levy |
The purpose of this paper is to present a SAS/IMLSOFIWARE program that can be used in conjunction with the NUN procedure in 1he SAS/STATSOFTWARE to calculate Box's bias approximation and Hougaard's measure of skewness. As an example, Box's bias approximation and Hougaard's measure of skewness were calculated on a leaf production data set (Ratkowsky, 1983; 1990) using the model Y=B1-B2B3 x* This program is compatible with Kugler and Lee's (1990) SAS/IML software program for curvature measures |
proc nlin, feature selection | source | 3 |
| 87 | Do We Always Need Clinical Stratification in Multicenter Clinical Trials? Results of a Simulation Study | Roland Brandmaier, U. Aydemir, H. Ansari, J. Hasford |
sample, sample size, relative risk, median survival, clinical trial | source | 6 | |
| 88 | Effective and Efficient Information Delivery for Executive Management | LeRoy Bessler | proc step, bar chart, vertical bar chart, proc print, proc printto | source | 7 | |
| 89 | Effective and Efficient Use of SAS/GRAPH Software | LeRoy Bessler | bar chart, proc step | source | 10 | |
| 90 | Efficient SAS Software Programming: A Version 6 Update | Jean Hardy | Most papers exploring efficient SAS programmi n9 have' been concerned by version 5. Surprisingly few addressed the efficiency of version 6 features: WHERE processing, indexing' and compression of SAS files, the SQL procedure, the use of engines. Some of these features decrease significantly programming and debugging time, although they can increase other resources. This paper repl ieates. with re 1 ease 6.06 J some eff i c i ency tests previously done with version 5 and adds many new ones targeted toward the new features |
dataset, proc step, where statement, put statement, cpu time, subsetting if, character variable, input statement, proc tabulate, proc sql | source | 5 |
| 91 | End-User Reporting Deliver More Information Sooner | Alan Eaton | The SAS" System provides a wide array of reporting tools. The SAS/ASSISP report button and the REPORT procedure offer both ease of use and powerful functionality. The information delivery strategies described here include exception reporting, drill down, and links to graphical charts using the REPORT procedure |
proc report, proc step, if-then-else, catalog entry | source | 9 |
| 92 | Enhancing the Printed Appearance of Your SAS Reports | S. David Riba | macro variable, put statement, variable | source | 3 | |
| 93 | Essential SAS Software for Engineers | William D. Kahn, Diccon R. E. Bancroft |
The main responsibility of engineers in industry is to economically produce consistent, high perfonning products. Because it is only through experiments that a product can he improved. one of the most important activities of engineers is experimentation. We have been teaching the principles and practice of experimentation to engineers at Gore for four years. We have isolated a necessary and sufficient subset orSAS which empowers engineers to 1: reflect on historical data, and 2: design, conduct, and analyze randomized, blinded, mixed-level fractional factorial experiments. This code takes about 16 hours of class time to teach. We now have over 100 engineers routinely doing experiments that include crossing and nesting, blocking, confounding, multiple error tenns, and ftxed and random effects. The statisticians at Gore are no longer directly involved with most of the designed experiments | proc step, proc sort, production run, proc print, sample, proc gplot, quality control, proc means, proc genmod, randomization | source | 10 |
| 94 | Estimating the Cost and Time for Developing SAS Software Applications | Joel M. Hoffman, Judith H. Mopsik, Jeffrey A. Polzin |
best estimate liability, clinical trial | source | 7 | |
| 95 | Evaluating Export Marketing Decisions Using Cluster Analysis and Correspondence Analysis | Marcus Schmidt, Jesper Strandskov, Gerald Albaum |
The study investigates the relationship between elements of an export marketing strategy and management style in marketing decision-making based on data of exporting companies from Oeomark, New Zealand, and USA. The deeision making styles were evaluated along seven dimensions (5-pliflt Likert scaled questions). A cluster analysis of factorscores (PROC PRINCOMP and PROC FASTCLUS) grouped the sample of firms Into four different management stylas per country. Discriminant analysis (PROC STEPDISC and PROe OISCRIM) was applied to test the robustness of the clustering. Considerable ditfSrQ[JC9S were founel between the decision-making sty/ss 01 the lour groups, At the same time remarkable similarities across countries could be obselVed when performing cluster analysis (PROC CLUSTER) on the 12 styles times 7 dimensions matrix |
sample, cluster analysis, discriminant analysis, proc corresp, confusion matrix, proc step, proc princomp, outlier | source | 5 |
| 96 | Experiences Using SAS/ACCESS Software in a Complex Relational Database Application | Murty Arisetty | The SAS System's Release 6.06 version of SAS/ACCESS software offers enhanced Relational Database to SAS System interface. RdbjVMSN (RDB) was selected for the application as a corporate relational database management system (DBMS). After extensive evaluation of commercially available Fourth Genera tion Languages (4GL's), SAS/ACCESS, SAS/AF and SAS/ASSISR software to gether were selected as the 4GL. The combination of the three SAS software products offered a long-term strategic solution. This paper discusses experi ences using SAS/ACCESS software and presents different tips, strategies, and techniques for developing appli cations using SAS/ACCESS software and |
relational database, data integrity | source | 9 |
| 97 | Experimental Design of the Mace with JMP Design | John P. Sall | For design and analysis of experiments SAS Institute is introducing a new product for the Macintosh, IMP Design. This product is a companion product to IMP, SAS Institute's statistical visualization product | source | 7 | |
| 98 | Experiments in Indexing | Swati Bansal, Margaret Seldin |
proc step, where statement, by group processing, lookup, put statement, cpu time, index creation, index | source | 3 | |
| 99 | Exploiting Features of Release 6.07 of the SAS System that Are Specific to the VMS Operating System | Brian Hess | dataset, proc step | source | 8 | |
| 100 | Extending the SAS System's Statistical Capabilities with SAS/IML Software | Charles B. Hallahan | Despite the fact that 'SAS' does not stand for you-know-what, it is true that between SAS/STAT", SAS/ETS', SAS/OR' and SAS/AC', the SAS System does provide an extensive variety of statistical procedures. However, there still are occasions requiring a statistical analysis for which a SAS PROC does not exist. At that point, IML, the Interactive Matrix Language, can be invaluable. This paper will give an overview of IML drawing on examples from bootstrapping, non-parametric density estimation and regression, Bayesian estimation imposing inequality restrictions, and graphics. In the discussion that follows, all IML statements and functions will appear in bold print |
sample, bootstrap, confidence interval, likelihood function, proc model, proc step, regression analysis, monte carlo, posterior probability, symbol table, options statement, maximum likelihood estimation | source | 10 |
| 101 | Faculty Productivity Information System | Mohamad Khachab, Dan Moser, Robert H. Waldman, Leon B. Ellwein |
Medical institutions have difficulty tracking faculty productivity. To address this problem, the University of Nebraska College of Medicine has developed a computerized database containing faculty profiles, publications, education / teaching contact hours, and clinical service patient contacts. The database was designed In 1985 with the initial purpose of providing rapid and complete communication of research endeavors to university administrators and outside constituents. SAS was selected at the time even though the project team had doubts that it was the best tool to handle this project. Lots of effort, and many versions of SAS code went into perfecting this system. Special direct access tables were built to reduce access time and Improve I/O processing. These tables have proven to be very useful and efficient. This method is still far better than using SAS V6 indexes. The system uses powerful but simple SAS techniques | call symput, macro definition, %if-%then-%else | source | 5 |
| 102 | Fun with the SQL Procedure - An Advanced Tutorial | Paul Kent | The SQL procedure is an implementation of ANSI Standard Structured Query Language (SQL) that facilitates extracting data from multiple sources with simple and concise queries. SQL is a versatile language for expressing these queries but there is often more than one way to pose the same question. This tutorial attempts to provide the reader with some insight into the SQL query optimiser and formulating queries that can be executed most efficiently. We hope some of the examples are sufficiently different from conventional solutions that you will gain increased awareness for the range of queries SQL can address |
proc sql, proc step, where statement, dataset, create table, macro variable, macro loop, macro definition, %if-%then-%else | source | 9 |
| 103 | Future Directions in Working with Procedure Output | Christopher R. Olinger, Robert S. Ray |
Users often express the need to access values that are printed by procedures but that are not available through an output statement. There is also a need to be able to customize procedure output in tenns of data arrangement and labeling. A current development project explores new alternatives for creating and processing procedure listings that will meet these needs and more. The Output Delivery System (ODS) gives the user greater access to procedure output than has been possible in the past. Procedures using ODS create persistent output objects from which you can make SAS data sets, produce printed listings, reorganize and save output in SAS catalogs, and create permanent custom report formats. Two experimental procedures in base SAS software enable you to manipulate and modify ODS output objects: the OUTPUT procedure and the TEMPLATE procedure |
proc step, sas output, proc template, proc mixed, catalog, output delivery system, dataset, if-then-else, restricted maximum likelihood, macro variable, likelihood function, proc genmod, proc glm | source | 8 |
| 104 | Generating Statistical Tables Meeting NDA Guidelines the Lazy Way | John Lue | proc tabulate, anova, least squares mean, proc means, proc step, degrees of freedom, proc genmod, proc glm | source | 6 | |
| 105 | Getting It and Keeping It with SAS Software-Advanced Strategies | Sally M. Muller, Kenneth A. Hardy, Keith J. Brown |
sample, dataset, sample size, cpu time, set statement, where statement, subsetting if, master data management, character variable, if-then-else, random sampling, by group processing, ranuni, view, proc sql | source | 7 | |
| 106 | Getting Mainframe SAS/GRAPH Output into Your PC Software | Timothy L. Latendress | Producing report quality graphics in a mainframe environment is extremely easy using SAS Version 6.06. However, the majority of desktop publishing tools are on PCs. How do you get your mainframe SAS/GRAPH output into PC programs such as WordPerfec~ Ventura Publisher, Lotus Freelance Plus, Harvard Graphics, or COREWRA WI? SAS/GRAPH version 6.06 supplies several device drivers for creating Computer Graphics MetafIles (CGM) that can be transferred to any program that imports the CGM fOlmat. Unfortunately, as anyone who has attempted to transfer graphics can attes~ it is not as simple as it sounds. This paper shows how to create a CGM for use in several popular software packages and provides notes on how to avoid many unexpected pitfalls |
proc step, error message | source | 4 |
| 107 | Getting the Most Out of the CMS Operating System with SAS Software | Matt Aaronson | Data access is improved in Release 6.07 of the SAS System for the IBM" eMS operating system with SAS/SHARE" and SAS/CONNECT' software, as well as access to VSAM and VSE files. Data management is improved with the provision of the TAPECOPY and T APELABEL procedures. Data presentation and data analysis are extended with further enhancements to support IBM 3270 devices. These features, along with the new interface to the 8AA REXX interpreter, performance improvements, and tips for tuning will help your productivity using the SAS System |
where statement, proc step, dataset | source | 6 |
| 108 | Graphical Aids for Summarizing Inferential Results of Nonlinear Regression | William J. Price, Bahman Shafii |
Nonlinear regression analysis requires iterative estimation procedures to achieve least squares solutions. Depending on the form and parametrization of the specified expectation function and the data being considered, these techniques may result in nonlinearity of the estimation situation with respect to one or more parameters. The degree of nonlinearity for each estimated parameter can be assessed by examining profile t plots, whereas the pairwise behavior of parameter estimates is examined using profile pair sketches. The extent of nonlinearity is determined by comparing these plots to those of the corresponding linear estimation situation. These plots may also be used to develop exact likelihood intervals and inference regions for the parameters. SAS" software codes to produce the required graphics are given and demonstrated using a two parameter exponential function | likelihood function, proc nlin, estimate, proc step, regression analysis, macro variable, macro definition, macro loop, %if-%then-%else | source | 6 |
| 109 | Graphical Methods for Categorical Data | Michael Friendly | Statistical methods for categorical data, such as loglinear models and logistic regression, represent discrete analogs of the analysis of variance and regression methods for continuous response variables. However. while graphical display techniques are common adjuncts to analysis of variance and regression, methods for plotting contingency table data are not as widely used. This paper provides a brief introduction to graphical | proc freq, proc corresp, proc step, logistic regression, odds ratio, data visualization, response variable, likelihood function, sample, regression analysis, proc princomp | source | 7 |
| 110 | Graphical Presentation of Gypsy Moth Severity Using dBase 111+ and the SAS System | Helen A. Machesky, Charles H. Coole, Daniel B. Twardus |
PROC MEANS was used to sunimarize the data below. The results indicate that total number of counties affected by gypsy moth defoliation has increased over the a-year period. The severity of defoliation within particular counties as indicated by the number of counties with more than 30 percent of their forest land affected. increased in 1990, then decreased in 1991. Year 1989 1990 1991 Total number of counties affected 163 214 222 |
proc step, proc format | source | 8 |
| 111 | Have It Your Way! Customized SAS Training: A Case Study of the Texas Instruments Training Solution | Suzanne S. Scott | Texas Instruments (TI), like many companies, does not have an unlimited training budget. With that in mind, a course was needed that would introduce the user to the SAS System as a whole, not just one specific part. While the SAS Institute offers many excellent courses, each covering only one product, TI needed a course that would cover several products and provide enough instruction to eliminate the need to attend (and pay for) multiple courses. A condensed, combined course was designed, through consultations with the Institute Training Staff, that Introduces 4 products: Base SAS, SAS/GRAPH, SAS/STAT and SAS/QC. Now, after only one course, students are comfortable enough with the SAS system (in general) to tackle any program. This approach allows teaching an |
sas base, bar chart | source | 3 |
| 112 | Hitting a Moving Target: A Simple but Effective Table Look-Up | Judy A. D. Alicante | macro variable, macro loop, variable, macro function, dataset, macro definition | source | 6 | |
| 113 | HOMOVAR: A Macro for Testing Homogeneity of Variances | Kristin Latour | In a one-way analysis of variance, variances are assumed equal across groups. When the data violates this assumption the AN OVA F test is invalid. This paper describes the SAS macro HOMOVAR that provides the following tests for homogeneity of variances: O'Brien Brown-Forsythe Levene Bartlett In addition the macro provides the Welch ANOVA test for means allowing unequal variances. All test statistics, degrees of freedom, and p-values are output to a user-<;pec~ied SAS data set | anova, f-test, degrees of freedom, response variable | source | 5 |
| 114 | How Southern California Gas Company Implemented an Interactive and Complex Forecasting System Using SAS Software | Tom Eunice | The Southern California Gas Company has had the Gas Margin Model since 1976. COBOL and FORTRAN programs and QSAM files were used to develop this model. Its function is to forecast the Cost of Gas and the Gas Revenue based on Gas Demand and Gas Supply forecasts. During the past fifteen years, this model has been modified and enhanced to accommodate regulatory changes. However. the requirements for changes have been increasing constantly and the model is no longer flexible enough to adapt to the new changes. Additionally, the nature of this batch system requires a great deal of user training. The model has been redesigned recently as an integrated on-line system to provide users with a friendly interface as well as the flexibility to accommodate the new regulatory changes in a timely fashion. SAS and DB2 have been used successfully in The first part the Cost of implemented in deliverables of this redevelopment. of the new system, Gas model, was April 1991 | proc step, proc sql | source | 4 |
| 115 | How to Gain a Working Knowledge of the FREQ Procedure without Freaking Out | Wendy London | The FREQ procedure will produce descriptive staJistics. i.e., n way frequency and crosstabulation tables, as well as perform inferential statistical analyses of Ihe relationships among variables. Version 6.06 statements and options needed to produce descriptive staJistics for printed output or output to a SAS data set are emphasized. No instruction in the use of PROC FREQ for inferenlial statistical analyses is provided. This paper is geared for the beginning to intermediate level SAS user. This will be * streamlined' instruction; not everything will be covered. but certainly enough to get started. The tutorial begins with a basic explanation of the PROC FREQ and tables statements. An explanation is given o/the descriptive staJistics output by the procedure. The handling of missing values, missing frequencies, and missing combinations with the missing, miss print, sparse, and list options is presented |
proc step, proc freq, by group processing, proc format, proc print, data step, missing value, proc sort, proc contents | source | 8 |
| 116 | How to Use the WHERE Statement | Juliana M. Ma | where statement, proc step, subsetting if, proc freq, cpu time, proc sql | source | 5 | |
| 117 | I/O Performance Improvements in Release 6.07 of the SAS System under MVS, CMS, and VMS | Stephen M. Beatrous, John T. Stokes |
where statement, cpu time, dataset, subsetting if, if-then-else, proc sort, bufsize | source | 5 | |
| 118 | Ifs, Ands, and Buts: A Case Study in Advanced Macro Implementation | Peter Kretzman | H you have chosen to delve more deeply into the macro facility of the SAS System, you probably started down the path by writing a few simple macros, and then noticing in the documentation that there are quite a few more capabilities than you are exercIsing. But, the intrepid journeyman macro programmer faces a variety of pitfalls: a bewildering set of abstruse quoting functions, multiple ampersand constructions, and a syntax which is just similar enough to regular SAS programming to get you into trouble. This advanced tutorial will piggyback on (and 8:ssume prerequisite knowledge of syntax-oriented material contained in) other previous SUGI tutorials on the macro facility, but will focus on elucidating the specific macro constructs and techniques ("idioms of expression") used by guru-level macro programmers | macro variable, macro facility, proc step, proc print, call symput, proc contents, flat file, macro definition, macro loop, macro parameter, sas expression, sas function, %macro parameters, %if-%then-%else | source | 7 |
| 119 | Importing SAS/GRAPH Output into WordPerfect | Bernarr Pardo | source | 5 | ||
| 120 | Importing X Window System Raster Images into SAS/GRAPH Software Output | Bernadette H. Johnson, Wendy D. Johnson |
proc step, conditional do loop | source | 6 | |
| 121 | Improving SAS System Support (Version 6) at a Large Site | Bruce F. Gilsen, Berton Shankman |
libname statement, autocall macro, catalog entry, macro definition | source | 8 | |
| 122 | Improving the Interpretation of Cluster Analysis: Developing a New Rescaling Algorithm | Soren Sten Hansen, Leif Thoisen, Marcus Schmidt, Ole Drengsgaard |
During the last two decades segmentation of cuslDmers has received increasing attention amongst the market research community. Usually, in such research some kind of clustering algorithm is used as an analytical tool for identifying life style segments. SAS Institute lnc.i& provides a number of clustering procedures based on different algorithms (i.e. linkage methods - single, average, complete, (two-stage)density - centroid, maximum likelihood, Lance-Williams flexible-beta., Gower's median, McQuitty's similarity analysis, Ward's minimum-variance, and Anderberg's centroid sorting). While some procedures like PRCe FASTCLUS provide statistical clues regarding the 'number-of-clusters' problem. no help is given with interpreting the operational significance of the identified cluster structure. Interpreting clusters can be a cumbersome if not impossible task which represents a serious obstacle to the successful practical implementation of the analysis | source | 6 | |
| 123 | Increasing Corporate Productivity with Decreasing Training Dollars | Jean Ussery, Bonnie Allen |
cpu time, sample | source | 4 | |
| 124 | Influence Measures in Nonlinear Regression | Tapon Roy | Nonlinear regression methods have gained great popularity in many different areas of application. The availability of software routines such as the SAS PROCEDURE NLIN has contributed to this increased use. Assessing the fit of the model and identifying possible outliers and influential observations are important in any regression analysis. Though many regression diagnostics are provided for linear models in the REG procedure, they are not provided in NUN. In this paper, several influence measures applicable to nonlinear regression will be described and a SAS program to calculate these measures and 'robust' parameter estimates is presented. An example of the methodology is also given |
proc nlin, regression analysis, estimate | source | 4 |
| 125 | INSIGHT into Multiple Regression | Robert A. Muenchen | do loop, regression analysis, proc step, bar chart, standard deviation, sample, estimate, macro definition | source | 4 | |
| 126 | Intelligent Production Graphic Reporting Applications | LeRoy Bessler | call symput, proc step, macro definition | source | 7 | |
| 127 | Interactive Data Analysis Strategies | Alvin M. Best III | data science, sample, response variable, regression analysis, missing value, null hypothesis, predictor variable, sample size, outlier | source | 6 | |
| 128 | Linear Time Lines: A Visual Summary of Critical Patient Data | Tenya Marie Economou, Pamela M. Hartigan |
The Veterans Affairs Cooperative Study Program developed a compact, yet comprehensive, display of critical data for each patient in an AIDS clinical trial. This display assisted statisticians in monitoring patient progress and included both the original treatment assignment and treatment changes. The display reports, at the time of occurrence, not only the following major study-related events: Progression to AIDS, Death, and T4Drop, but also the Drug Resistance monitoring data. We used the base SAS'" System and the SAS Macro Language Facility to produce linear time lines for each patient. While elegant in their simplicity, the time lines give a complete summary of patient progress in a concise, visual format. This paper describes the techniques used to create the linear time lines. Operating Systems: MVS,CMS,PC-DOS,MS-DOS | sample, put statement, character constant, pointer control, %if-%then-%else | source | 6 |
| 129 | Little Orphan Annotate: How to Dress Up SAS/GRAPH Software Output | Arthur L. Carpenter | The ANNOTATE Facility within SAS/GRAPH has a reputation for causing new graphics users headaches and frustration. Many abandon hope of making use of this important graphics tool. Much of this confusion is caused by the approach that ANNOTATE uses to gather the information it needs to place text, lines, and symbols on the plot or graph. ANNOTATE is not like much of the rest of SAS . Directions are not passed to the ANNOTATE Facility using a PROCedural call. Specific statements are not used to define parameters. Instead, information is passed to a PROCedure through a specialized SAS data set. To make matters a bit more arcane, the user of the ANNOTATE data set is not free to select variable names or their attributes. Although this is very contrary to both the traditional DATA STEP and the PROC STEP, it does not mean that using ANNOTATE needs to be difficult |
proc step, dataset, proc gplot | source | 6 |
| 130 | Load Data Analysis Workstation (LDAW): A SAS/AF Information Systems Application | Burtt Blodgett, John Powers |
data cleaning, macro variable, dataset, data science, proc step | source | 6 | |
| 131 | Macros Invocation Hierarchy: Session-Compiled, Autocall, and Compiled-Stored Macros | Susan M. O'Connor | This paper describes the compiled stored macro facility's function ality and syntax. A comparison of positive and negative consider ations in selecting the use of session compiled, autocall, and compiled stored macros is made. Such issues as maintenance, per formance, and development environments are discussed. Recommendations for macro-related considerations, such as auto call concatenation versus permanent macro catalogs and macro option defaults, are discussed. For a greater in-depth overview, related issues of hierarchy for these types of macros, of macro vari able resolution, and of the interaction of compiled stored macros and the stored program facitity are discussed. The timing and resolution of CALL EXECUTE and the RESOLVE function with compiled stored macros and session stored programs are addressed. An overview of these facilities and the order and tim ing of them might be helpful for users | stored macro, macro facility, macro catalog, autocall macro, macro definition, libname statement, catalog, macro variable, options statement, cpu time, flat file, macro call, quality assurance, call execute | source | 5 |
| 132 | Making Use of the SAS Implementation of SQL | Don Stanley | proc sql, proc step, create table, inner join, where statement, dataset, proc means, group by, proc sort | source | 6 | |
| 133 | Managing General Clinical Research Center (GCRC) In/Outpatient Census Data Using the SAS System and DEC VAX VMS DCL | Jacqueline A. Wendel | cpu time, proc step | source | 5 | |
| 134 | Managing Time and Resources: Enterprisewide Project Management | Laurie Rose | Project management is no longer limited to the traditional project industries, such as· construction and engineering. The abundance of information and the increasing complexity of projects have increased awareness of the need for tight management and control. As a result, project management software is being used in virtually every important business area: utility, aerospace, defense, financial, government, pharmaceutical, software development, automotive and manufacturing industries. By integrating information relating to activity progress, manpower, materials, cost and time into presenta tion-quality reports, project management software provides manag ers with the ability to improve their understanding 01 project status and to speed decision making-the key to successful project man agement | linear programming, sensitivity analysis | source | 4 |
| 135 | Mapping Alcohol Availbility in Los Angeles County: A Practical Application of SAS/GRAPH | Richard A. Scribner, Jon Galaif, David P. MacKinnon |
The purpose of the Local Alcohol Availability Database (LAAD) is to assist in the development of alcohol policy in Los Angeles County at the community level. Interactive maps produced with SAS/GRAPH can present an extensive amount of LAAD produced data in a single visual presentation. The unique characteristics of Choro, Prism and Surface maps are ideal for conveying information related to the physical availability of alcohol quickly and efficiently. This type of data presentation is essential for informing non-statistically grounded policy makers |
proc gmap, proc step | source | 5 |
| 136 | Marginal. Probabilities: An Intuitive Alternative to Logistic Regression Coefficients | P. M. Wright | This paper presents an algorithm for using output from PROC LOGISTIC to compute marginal probabilities. Logistic regression coefficients are difficult to interpret, because they reflect the effect that a change in an independent variable would have on the term In(P/(I-P)) - where P is the estimated probability that the dependent variable is one. Marginal probabilities, on the other hand, reflect the effect that a change in an independent variable would have directly on P. Because marginal probabilities are more intuitively interpretable than logistic regression coefficients they can be a more useful statistic. PROC LOGISTIC does not provide marginal probabilities directly, but it does provide everything needed to compute them. In troduction Statistical software packages such as SAS have made producing sophisticated statistical results as easy as 1[. This has been a tremendous boon to research |
logistic distribution, logistic regression, predictor variable, regression analysis, response variable, proc step, proc logistic | source | 3 |
| 137 | Marketing Research: Uncovering Competitive Advantages | Warren F. Kuhfeld | The SAS@ System provides a variety of methods for analyzing marketing data including conjoint analysis, correspondence analysis, preference mapping, multidi mensional preference analysis, and multidimensional scaling. These methods allow you to analyze pur chasing decision trade-off's, . display product position ing, and examine differences in customer preferences. They can help you gain insight into your products, your customers, and your competition. This paper discusses these methods and their implementation in the SAS System | conjoint analysis, proc princomp, proc corresp, proc step, nominal variable, data science, predictor variable, anova, regression analysis, proc freq | source | 9 |
| 138 | Maximizing Computer Resources with Cooperative Processing and Distributed Data Access | Jack Wallace | source | |||
| 139 | Model Selection for Uncensored Lifetime Data Using SAS/IML Software | Sutaip L. C. Saw | A SASjIML program is presented which enables one to select a best fitting model given a random sam ple of uncensored lifetime data. The competing mod els include the two-parameter gamma, inverse Gaus sian, lognormal and Weibull distributions. The program computes maximum likelihood estimates and selection statistics under each of these models. Two different types of model selection criteria are used. The best fit ting model, according to each of these criterion, is one whose selection statistic is smallest | feature selection, normal distribution, likelihood function, lognormal distribution, proc step | source | 5 |
| 140 | Most Creative Use of the Software | Nelson R. Pardee | The enclosed graphs attempt to illuminate the underlying grammatical and syntactic structure of New Testament Greek. In particular, displaying these language elements of the text helps to highlight usage patterns both of the language and the author. History The data and general techniques are based on work done by Edward A. Robson for a dissertation in linguistics at Syracuse University during the 1970s. He has continued the work as a personal research project; I became involved in 1990 to help with SAS programming and output design. Data The data used to create the graphs has one observation for each word in the text. Each observation currently has variables indicating chapter, word number (consecutively numbered from chapter beginning), part of speech, verb tense (if applicable), preposition information (if applicable: preposition, case, and prepositional phrase duration), and relative clause information (similar to prepositional phrase) | proc step, data step | source | 8 |
| 141 | Moving a Health Data System from a Research to a Business Environment | Michael C. Hein, Lisa Hein |
Health research groups and businesses in the health care industry have an understandably different set of goals and methods. This is especially apparent when comparing the systems development styles and overall computing needs of the two groups. When converting a data system that was born in a research project into a usable tool for a corporation, moving the data from the university computer to the company machine is only one step among many. The priorities of the researcher are not those of the business person. This difference creates tasks such as paring away hundreds of intermediate research files, enhancing the code and data structures to provide new output or interfaces (reports, export files, and so on), and, of course, documentation. One client of ours, a medical systems firm, needed just this conversion when they acquired a data system of intensive care unit records from a university research group | dataset, proc step, regression analysis, data cleaning | source | 5 |
| 142 | Moving SAS Transport Files Across Different Hardware Platforms - An Advanced Tutorial | Tamara R. Fischell, Margaret Menache |
The availability of numerous hardware platforms due to the growing affordability of computer equip ment and the essential need to use the most efficient tool, has propagated a multi-hardware architecture in the workplace. Not only are programmers forced to program on many systems in-house, but they must also port data to clients, who mayor may not have compatible systems. Moving information across different platforms, consequently, becomes key to ensuring interconnec tivity. SAS/CONNECT is one way to move SAS datasets, but when versions 6.06 (and higher) replace lower SAS versions, CONNECT must be licensed to use PROCs UPLOAD /DOWNLOAD. Transport files, another way to move datasets, will then rival CONNECT; not only because they will be more eco nomical, but also because they are now the only way to move SAS data across PC Local Area Networks (LANs). 1 Presently, there exists little integrated documenta tion on the total process of creating transport files |
proc step, dataset, proc copy, proc contents, proc cport, sas library, select, sas syntax | source | 8 |
| 143 | Multicriteria Decision Making with SAS Software | Mary G. Crissey | In today's world, many problems can not be optimized via a traditional linear program such as provided with SAS/OR software via PROC LP. Recently in the area of multi-criteria and multi-objective modeling many new algorithms have been designed to aid decision makers who deal with real world problems. The unique Interactive Sequential Goal Programming (ISGP) procedure was implemented in SAS to help military enlisted force mangers face many competing goals across multi-year time horizons. This paper shows how the basic ISGP algorithm can be applied in a generic sense with other SAS databases. 471 A short summary of the ISGP |
sample, linear programming | source | 7 |
| 144 | Multitrait Scaling Program: MULTI | Ron D. Hays, Eric Yu I. Wang |
hem convergence within mutti-item scales, or internal consistency, is commonly evaluated using Cronbach's (1951) alpha reliability coefficient. Evaluating item discrimination across scales, multitrait scaling analysis, is also desirable (Hays & Hayashi, 1990). The purpose of this paper is to describe a program to compute the necessary statistics for muhHraH scaling using the SAS/STAT" software and the SAS· macrofacilHy |
correlation, proc step, proc univariate, standard deviation, macro definition | source | 6 |
| 145 | Multivariate Inequality Hypostheses Using SAS/IML Software | Dan Jacobs, Estelle Russek-Cohen |
Muttivariate inequality hypotheses and muttivariate one sided tests have received considerable attention in the statistics literature (e.g., Barlow et aI., 1972 and Shapiro, 1988). However, they are often avoided by practicing statisticians. The lack of readily available software may be a major obstacle in the use of these methods. While FORTRAN algor~hms have been published for various pieces of the analyses required, no integrated package for these procedures exist. SASlfML software modules have been developed that are modificationsolthe established FORTRAN procedures. These modules, along with the flexible programming features of SASlfML software, make the testing of a number of muttivariate inequality hypotheses fairly straightforward | likelihood function, null hypothesis, sample, regression analysis | source | 4 |
| 146 | New Features in SAS/TOOLKIT Software | Richard D. Langston | Release 6.07 of the SAS System under MVS, CMS, and VMS operating systems is the initial produc~ tion release of SASjTOOLKIT software for those same operating systems. This paper discusses some of the features that have been introduced with this pro duction release since the experimental release in July 1990. The major emphasis of the paper is on the new engine writing feature | dataset, symbol table | source | 7 |
| 147 | New Publications for SAS Software Users | Wanda Verreault, Anne Corrigan |
The SAse System is the world's premier information delivery sys tem. As an integrated applications system, the SAS System gives you complete control over data access, management, analysis, and presentation. With the SAS System, you can perform data entry and retrieval, as well as report writing, graphics design, statistical and mathematical analysis, business forecasting, operations research, quality assurance, and applications development. Some people use many of the SAS System's capabilities; others use only a few. But no other software gives you so many choices for meeting your orga nization's information needs. This paper summarizes some of the ways the Publications Division of SAS Institute Inc. is continuing to build your SAS documentation library, reviews new publications from the past year, and previews upcoming offerings. It also discusses the Institute's Books By Users program· | sas base, proc report | source | 3 |
| 148 | New Technology - Transforming a Mainframe System to the PC Using SAS/AF Software | Henry F. DeLong Jr., Valerie J. Martin, Beth A. Schlaline |
The PC has become an almost indispensable tool in the workplace today. It has only been recently, however, that pes have become large and/or powerful enough to take over processing tasks previously slated for the less convenient mainframe environments. This paper will outline th~ steps necessary to successfully convert a mainframe processing system to a PC environment capable of handling data from a variety of sources including ASCII. EBCDIC. dBASE~, and Lotus 1-2-3'". Certain questions need to be answered whenever changes are made to an existing production system. Will the changes save valuable resources? Will the changes enhance the reliability of the data? * How will the user adapt to the changes? These questions become even more important since our available technology is changing so rapidly. Examples of converted code will be given as well as suggestions for integrating the existing procedures into a menu driven system using SASIAF | proc step, data wrangling, cpu time | source | 5 |
| 149 | Optimal Clinical Data Structures for Use with SAS/PH-Clinical Software | Sandra D. Schlotzhauer, Andrew T. Fagan |
SAS/PH-Clinical software is a new SAS Instnute software product for increasing the productivity of clinical reviewers within drug or biotechnology research organizations. Atthough the product is designed to work with a variety of clinical data structures, the internal algorithms handle some structures more efficiently than others. In addnion to data set structures, any aspect of the clinical data that requires a view overlay affects the efficiency of the software. For example, n variables must be renamed in the study definnion process, SAS/PH-Clinical software overlays a DATA STEP view on the original clinical data. Before any patient group can be created, the software generates this view and reads n, rather than reading the original data. This paper discusses aspects of clinical data structures that can affect performance |
date variable, clinical data, laboratory data, full join, lookup, cpu time, missing value, adverse events | source | 8 |
| 150 | Optimal Tax Strategies for Small Businesses and Professional Practices | Janet R. Gibley, Raymond A. Gibley |
Some of the principal causes of failure among the twenty million small businesses in the United States are poor financial management and overpayment of taxes. This particular paper describes how the small business owner and professional can minimize federal and state income taxes consistent with a company's cash flow budget. The authors characterize the tax planning problem as a 0- 1 knapsack problem and employ PRDC LP to optimize the overall tax burden. The model is developed using tax la\/\!._ from the Internal Revenue Code of 1986 and all modifications since then. These tax planning strategies have been discussed on the evening news on KMGH-TV, the CBS affiliate station in Denver, Colorado | health insurance, sensitivity analysis | source | 3 |
| 151 | Ordination and Display of Multivariate Data | E. Barry Moser | Graphical displays of mUltivariate data are a standard part of multi variate analysis and modeling. Multivariate ordinations and dimen sion reduction techniques permit low-dimensional views into higher dimensional space giving insight into model fitting and inter pretation, and to the presentation of data and results. Minimum spanning trees, biplots, and other devices further expand these capabilities. The PRINCOMP and CORRESP procedures of SASI STAT", and SAS/IML and SAS/GRAPH software, enable analy sis and display of continuous and categorical multivariate data. Examples using real data are used to illustrate these techniques |
macro definition, scatter plot, proc corresp, proc princomp, correlation, covariance, proc step, proc freq, distance measure, manhattan distance, sample, outlier | source | 10 |
| 152 | Performance Enhancements in SAS Procedures for Small and Large Applications | Leigh A. Ihnen | The term Release 6.07 will be used to in dicate only the current Release 6.07 of the SAS System for MVS, CMS, and VMS. Re lease 6.07 continues the enablement of small and large applications through performance improvements at the procedure level. Solu tions for applications too large to run in Re lease 5.18 are enabled by exploiting larger address spaces. At the same time enhance ments to procedures allow small jobs to run faster. The continuing evolution of Version 6 of the SAS System aims to provide cost effective solutions for a wide range of enter prise applications. With this release better utilization of host-dependent features such as IBM XA and ESA architecture, vec tor facilities, and interfaces to other soft ware allow applications to be designed on a larger scale | proc step, proc sort, dataset, cpu time, proc format, raw data, proc genmod | source | 10 |
| 153 | Pharmacokinetic Statistical Analysis System - A SAS/AF and SAS/FSP Application | Sharon Passe, Andrea L. Contino |
The statistician responsible for the design of pharmacokinetic studies, the subsequent analy sis ofthe data and the report generation, typically repeated the same time-consuming tasks for each new study. A menu-driven system was needed automating these routine functions. The Pharmacokinetic Statistical Analysis System (PSAS) is a SAS based system created to expe dite design tasks, data processing, statistical analysis and report generation for the pharmaco kinetics area. Base SAS, SAS/FSP, SAS/AF, SAS/STAT and SAS/GRAPH (Version 6.07) software are used in this workstation based system. Modifica tions and the incorporation of additional options are easily made because of the flexibility of SAS/ AF, SAS/FSP and Screen Control Language |
dataset, sample, anova, sample size, proc glm, proc means, sas base, data science, proc genmod, proc step | source | 6 |
| 154 | Picturing Production Data Using Base SAS and SAS/GRAPH Features | Tamara L. Siler | As part of regular reviews, we must be able to present large amounts of production data in an informative, timely fashion. To assist us in this we have developed an application using data management and graphics techniques offered by Base SAS and SAS/GRAPH. Several procedures will be discussed, including PROe FORMAT, PROe GPLOT, and PROe GREPLAY. In using these features, we have truly made our "picture worth more than a thousand words" |
proc step, proc gplot | source | 5 |
| 155 | Pie Charts and Bar Charts: Getting Their Best out of SAS/GRAPH Software | LeRoy Bessler | bar chart, line chart, proc step | source | 10 | |
| 156 | Planning for the Future with the SAS System | Robert S. Lohr, Robert Nunziata |
source | 3 | ||
| 157 | Plot Mean with Standard Error by Using the SAS/STAT GLM Procedure and the SAS/GRAPH GPLOT Procedure | Eric Yu I. Wang, Aili Gong |
Plots of means with their standard errors are widely used in statistical presentations. The purpose of this paper is to describe a program to automatically use PROC GLM to get adjusted mean and standard errors to use within the SAS GRAPH GPLOT procedure and the SAS MACRO program. The use of the program is demonstrated in an example of our prevention research |
proc calis, proc step, proc glm, proc gplot, proc genmod, macro definition | source | 6 |
| 158 | Plotting with Mainframe Power and PC Flexibility in Version 6 of the SASSystem | Tim Berryhill | proc step, macro variable, do loop, macro quoting function, where statement, data wrangling, proc gplot, usage data, %if-%then-%else, macro loop, macro definition | source | 5 | |
| 159 | Point and Click Your Way to a Mainframe Database: The Easy Way to Download Data to Your PC | John Roback | This paper describes an appltcation written using SASlAF software that makes downloadiOQ SAS data from a mainframe fast and easy. This PC-to-mainframe Interface application allows users to select subsets of a mainframe database and download those subsets in a fonn useable by other PC- and Macintoshll!>-basad applications. No knowledge of SAS software or mainframes is required. Using a mouse, users simply point-and-click on the database they want and the software hand1es the delais of locating the daIa on the mainframe. Statistical and graphing procedures are available to aid users in selecting a subset of the database to download. Users can download the data in a format compatible with most spreadsheets, word processors. and statistics pacl<ages. The paper emphasizes the tools and Ied1niqJes usad to ""ate the app6cat1On and how k is used within a research environment | flat file, sas output | source | 5 |
| 160 | POWER: A Simple Macro for Power and Sample Size Calculations | Kristin Latour | Statisticians often desire to investigate the power of statistical effect tests, or to estimate the sample size required to provide a significant effect test. This paper describes the macro POWER which provides the following power related measures: Power of a test Adjusted power and confidence lim~s Least significant number The POWER macro accepts a PROC GLM OUTSTAT data set as input. Users can supply an unlimited number of significance level, sample size, standard deviation, and effect size values. The calculations listed above are performed for all combinations of these parameters and output to a SAS data set. The macro handles prospective as well as retrospective power analyses |
sample, sample size, effect size, power analysis, proc glm, proc genmod, significance level, mse, proc step, standard deviation, confidence interval | source | 5 |
| 161 | Processing Longitudinal Data for Input into the LIFETEST Procedure: Examples from the Survey of Income and Program Participation (SIPP) | Taube Wilson | The Survey 01 Income and Program Participation (SIPP). a panel studV condIctad by the Bureau 01 the CenstJa. Is a """""">us resource tor economists and social scientists in government and _ alik . ThIs unique source 01 data has played a role in public policy analysis oovering civersily 01 IDpIca .uch as wallar. rocipioncy. 1abor force participation. and small business ownership. Because 01 its siz. and eIaborale file s_tur ** the SIPP pres ...... a particular challenge ID the program""". SAS Is ideally su_ for this application. In which latgo amoun .. 01 complex data must be aggogalad at _ .... t levels for statistical processing. ThIs _ COY .... I8chnlquos for constructing monlhly household level agg~ and 1Iags. methods for creating variables used for IongitucInaI spall analysis. and I8chniques for reorganIZing the data ID IscHits'" these proces | proc step, right censoring | source | 6 |
| 162 | Project Tracking System Using SAS/AF Software | David M. Hartman | dataset, macro variable | source | 6 | |
| 163 | Proportional Hazards Analysis of Left-truncated Survival Data | Dennis Heisey | Failure time data are typically recorded as the times from some origin until some endpoint, such as the curing of a disease, or death. The influence of covariates or prognosticators on time until endpoint is often 01 interest. The proportional hazards model (Cox 1972) has become a very popular approach for evaluating the influence of covariates on failure times. PROC LlFEREG and PROC PHREG are examples 01 procedures that perform proportional hazards modelling 01 time until endpoint data. Such procedures assume that all subjects effectively enter the study at the same time as the origin of the process that gives rise to the endpoint of interest. This is not an appropriate assumption for some studies. For example, in studies of infection incubation times, it is_ not uncommon lor subjects to enter the study well after initial infection {the origin} if initial infection is asymptomatic |
proc step, likelihood function, proc phreg, regression analysis, sample, survival analysis, hazard function, logistic regression, binomial distribution | source | 6 |
| 164 | Put Presentation Graphics in Your Applications | Anthony L. Friebel, Jade Walker |
In developing applications using SAS/GRAPH'" software, perhaps you have come up with some clever tricks to enhance the appear ance of your graphs. This paper presents a mixed bag of tricks col lected over the years. The authors have tried to streamline the code examples and to show how to generalize display enhancements so that you do not have to code a separate program for every input data set you use. Techniques employed in this paper include the use of other parts of the SAse System to gather information from your input data set, the use of MACRO variables to make global changes over a program, and the building and subsequent including of external files containing SAS statements with parameters that are automatically based on your Input data set. Using these general ized algorithms reduces the amount of SAS source code that you have to maintain, and, who knows, this paper may even include as an example the very graph you have been trying to create! |
proc step, bar chart, footnote statement, macro variable, dataset, proc freq, proc gchart, proc means | source | 10 |
| 165 | Quality Sciences at Delco Electronics Corporation | Bruce McLean | proc step, proc glm, proc genmod, sample | source | 3 | |
| 166 | Realizing Improved Efficiency on Large Databases with SQL in SAS 6.06 with Further Improvements under SAS 6.07 | Anne Soeder | Attempting to JOIN a small SAS dataset with a large OB2 table using PROC SQL can require large amounts of memory and CPU time. This JOIN can be made more efficient with a macro which will soon be available in the SAS.SAM PLE.LIBRARY Further improvements can be realized with improved SQL processing in SAS version 6.07 with the Pass through feature. This paper will demonstrate how the use of one or both of these techniques will reduce up to seven fold, without increasing memory requirements |
cpu time, dataset, proc sql, proc step, create table | source | 3 |
| 167 | Resource Usage and Capacity Management in Large UNIX Environments | Robert K. Newman | What has been standard resource usage and capacity management practice for most mainframe shops is not yet in common practice in many UNIX based environments. With the continued movement of business "critical" IS functions to UNIX based distributed processing systems, the current situation is changing. This paper recommends a UNIX resource usage data collection process that satisfies the data requirements of UNIX capacity management. It also addresses the problem in large UNIX environments of data overload, by recommending a specific set of key indicators to retain, as well as what UNIX as and network resource usage levels to use for exception reporting. The recommendations are based on the experiences of working in Northwest Airlines' large, highly integrated UNIX environment used for the processing of several important IS business applications | usage data, capacity planning, distributed processing, network utilization, data governance | source | 7 |
| 168 | Running the Business: Financial Management with the SAS System | Jeff Babcock | The theme of SUGI17 is Information Delivery. My goal is to discuss its application to the Financial Management process, which is the generation of strategic information necessary for the achievement of business goals in an organization | data science, data governance | source | 14 |
| 169 | SAMP/AID - Sampling Applications and Demonstrations in Learning Statistics | Efraim Darom | Sampj Aid is a menu driven application useful as a supplement for courses in Statistics. End users pick items from a menu, and specify parameter values which are read by SAS macros. Some of the macros produce sample data files which can be further analyzed. SAYIP/AID is supported by IBM's VYliCYlS and its operation does not require knowledge of SAS programming | sample, correlation, dataset, sample size, effect size, regression analysis, standard deviation | source | 5 |
| 170 | SAS Classes for Nonprogrammers in the Clinical Data Management World | Susan V. Hale | Computers are becoming an every day part of everyone's lives and SASID has made life in the C Hnieal D ata Management world easier. Utilizing SAS BASICS, Clinical Data Specialists can program simple SAS queries in order to more effectively perform quality assurance tasks on clinical data. Developing a basic programming skills class reviewing such things as DATA steps, IF· THEN·ELSE logic, SET, MERGE, SAS dates, PROC SORTS, and PROe PRINT logic, we in COM have turned our non programmers into efficient basic SAS programmers which has resulted in quicker and more effective data quality assurance |
set statement, quality assurance, merge statement, sample, sas base, sysdate9, proc contents, adverse events, data cleaning, proc step, clinical trial, outlier | source | 4 |
| 171 | SAS Curriculum - An Integrated Approach | Patti Zehner, Lynda M. Randall |
source | 6 | ||
| 172 | SAS Software - The Power Tool for Information Professionals | Shane Wood | Information is Power. We are in the midst of the Information Age, the period of most rapid technological change since the Industrial Revolution. Information is the life-blood of any organisation. Accurate and timely infonnation can provide significant, competitive advantages to a company, and incorrect or late information can destroy it overnight! An essential requirement for informed decision making in business, and life in general, is prompt access to accurate, consistent and credible information. Many different products are available for the storage, management, access and presentation of information, but there is only one that provides both skilled and unskilled users with a completely integrated solution embracing all these areas and many others. This product is SASRl . The vast volumes of information generated in a modem organisati?" also create the new problem of information overload | data science, key performance indicator | source | 5 |
| 173 | SAS Software and the Bard: Using SAS Software Procedures to Measure One-Person Theatrical Show Preferences | James D. Ryan, Joseph E. Earley |
proc step, proc reg, proc freq, regression analysis | source | 6 | |
| 174 | SAS Software on the IBM RISC System/6000 Workstation - The Wave of the Future for Information Delivery | Darius S. Baer, James L. Shefelbine |
After years of using the SA$' System for a wide variety of applications on both MVS and VM mainframe platforms, many SAS users may feel somewhat intimidated by moving to the IBM RISC System160()() (RS/6000) workstation environment. Actually, once the users become somewhat familiar with the AJ)( operating system (IBM's UNIX offering), they will feel rather comfortable. We discovered that running the SAS System on the workstation was not only similar in many ways to running on a mainframe, but it was actually easier. The availability of the graphical interface and locally attached terminal made the SAS experience both more interesting and enjoyable as well as faster in most cases. The AIX environment offers unique opportunities and challenges for an organization heavily dependent upon the mainframe environment. Historically, the Burlington Northern Railroad has developed SAS applications solely on the mainframe | data wrangling, dataset, batch job, batch processing | source | 6 |
| 175 | SAS Software Puzzles 1992: Conversion Techniques | Veronica Fasullo, Robert G. Hall |
macro variable, macro definition, macro put, symbol table, comment, local symbol table, proc step | source | 10 | |
| 176 | SAS Users Groups: A Network of SAS Software Professionals | Sally A. Roberson, Betty Fried |
For the past three years, SAS Institute has had a formal program for users group development and support. We actively support groups at the local, regional. national, and intemationallevel. While the Institute supports SAS 8 users groups worldwide, this presenta tion will look at our program for groups in North America | group by, excel | source | 3 |
| 177 | SAS/EIS Software: Developing Applications for Information Delivery | Gail Kramer, Jamie Robbins |
dataset, vertical bar chart | source | 3 | |
| 178 | SAS/IML Software: An Application-Oriented Introduction | Ed Barber | dataset, proc iml | source | 4 | |
| 179 | SAS/NVISION Software Graphics | Anthony R. MacKay | source | 2 | ||
| 180 | SAS/NVISION'"Software: The Sequel | Anthony R. MacKay | source | 1 | ||
| 181 | Simple SAS Analyses That Add Value to Corporate Databases | Richard Sue | proc step, dataset, missing value, numeric variable, proc tabulate, proc freq, record count, proc means | source | 4 | |
| 182 | Small Area Analysis of Hospitalization Rates Using SAS Software | Larry J. Walter, Bruce H. Fireman |
The Kaiser Permanente Medical Care Program in Northern California, like other health care providers during the current period of rising medical costs, has a pressing need to manage its hospital utilization while maintaining a high quality of medical care. Small Area Analysis (SM), which applies population-based statistical methods to compare hospitalization rates across hospital service areas, can be helpful in identifying high-variability medical diagnoses and procedures where utilization management may be effective. The SAS System application described here produces detailed graphical reports for comparing hospital utilization rates for many diagnoses and procedures across different hospaals, time periods and populations. SASjGRAPH· was chosen for the production of these reports because of its ability to produce a sequence of high quality customized graphs under. control of the macro language | proc means, likelihood function | source | 6 |
| 183 | Smarter SAS Programming through the Use of Coding Conventions | Linda A. Smith, Anne M. Asher |
macro put, macro quoting function, proc step, sas log, macro definition, macro variable | source | 6 | |
| 184 | Some Considerations in Designing SAS-Based Data Management Systems for Large International Epidemiologic Studies | Cynthia L. Johnson, Michael Arrighi, Priscilla E. Van Grevenhof, Roger Mueller, Laurel Panser, Robert Lee, C. Chute |
Special considerations apply when processing large databases in a PC SAS environment. A SAS-based data management system was developed to handle data stemming from a large international epidemiologic study to investigate the natural history of benign prostatic hyperplasia. Over 4000 men aged 40-79 are participating from Olmsted County, MN and Scotland Stirling district communities. Follow-up visits are planned. An eventual goal of combining data from these transatlantic sites focused initial efforts on comparability of databases, resulting in the exclusive use of SAS for both data storage and data processing. The massive amount of data collected via self-administered questionnaires, urinary flow rate devices and clinical evaluation along with the desire to retain processing on microcomputers forced considerations of efficiency in designing databases. Procedures for data flow, including data transfer, backup and error correction were developed along with approaches for developing | dataset, data wrangling, data science, proc step, proc sql | source | 9 |
| 185 | Some Problems in Computer Aided Design of Experiments | Douglas C. Montgomery | The widespread availability of computer hardware and software to support statistical design of experiments has made it possible for experimenters to easily use this powerful approach to problem solving. Computer software is useful in automating some aspects of the administration of an experimental design program, and performing the data analysis. In recent years, computer programs to assist the experimenter in generating a design for a particular problem situation have become more widespread. Some of the situations in which these programs are used include irregularly-shaped experimental regions, such as in a mixture experiment, or response surface modeling problems in which there are supplemental conslraints. We discuss and illustrate some practical aspects of using these computer-generated-design aigorithms, and show how care is required in order to produce an appropriate experimental design in most situations | uniform distribution, multicollinearity, degrees of freedom, regression analysis | source | 8 |
| 186 | Some Suggested Modifications to the SAS Conjoint Macro_ | Jairo Castano, Willem Janssen |
cpu time, correlation | source | 6 | |
| 187 | Sources of SAS System Information in the MVS Environment | Michael A. Raithel | cpu time, sas library, batch job, options statement, elapsed time, proc contents, sas programmer, proc step, proc options | source | 10 | |
| 188 | Spin, Rotate, and View Three-Dimensional Point Clouds: Dynamic Exploration of SAS Data Sets without SAS/INSIGHT "Software | Arthur L. Carpenter, Clinton A. Carpenter |
numeric variable, set statement | source | 6 | |
| 189 | Statistical Simulations Using the CALIS Procedure | David P. MacKinnon | The purpose of this paper is to describe how to conduct statistical simulations of covariance structure models using SAS BASE, SAS STAT and the new CALIS program. Other covariance structure programs like LISREL (Joreskog & Sorbom, 1989) and EQS (Bentler, 1989) have statistical simulation capabilities. In this respect, CALIS is not as useful as these other programs. However, as shown below the SAS program can be used to conduct general statistical simulations for covariance structure models. The program consists of generating data based on the regression equations for the hypothesized model in the DATA STEP, estimating the model in CALIS and then repeating this step for a number of replications to determine the sampling variability of simulation statistics. The |
covariance, proc calis, covariance structure, monte carlo, sample, sample size, proc step, regression analysis | source | 5 |
| 190 | Strategies for Advanced Merging Applications Using the SQL Procedure and the Data Step | Judith H. Mopsik, Jeffrey Phillips, Margaret Seldin |
proc step, group by, proc sql, create table, by group processing, balance, cpu time, order by | source | 8 | |
| 191 | Tabulating the Results: An Application of SAS Software to Periodic Reporting in a Longitudinal Epidemiologic Study | William C. Murphy, Howard M. Proskin, Dennis H. Leverett |
Reports consisting of dozens of similar tables are required for the periodic monttoring of an ongoing epidemiologic survey. Data gathered in the field are regularly assembled into the SAS system for study and analysis. PROC TABULATE. with tts various options, is ideal for producing the tables needed for the periodic reporting of stUdy |
proc step, proc tabulate, macro definition | source | 4 |
| 192 | Taking Advantage of the Power and Flexibility of the SAS System for OS/2 and Microsoft Windows | Mark W. Cates | The two preeminent operating systems for personal computers, OS/2 Version 2.0 and Microsoft Windows 3.1, are rolling out new releases, upgrading their power and flexibility_ The two operating systems provide a solid platform for SAS users on workstations. However, the two systems present radically different architectures to the SAS System. The operating systems differ in the following features: memory architecture, flat versus segmented; multitasking, preemptive versus cooperative; processing power, 32·bit versus 16-bit; file 1/0 performance, HPFS versus the FAT; DOS compatibility; and user interlace. 80th operating systems provide an array of interprocess comm unications methods allowing dynamic sharing of data between applications. Examples are Dynamic Data Exchange, Object Linking and Embedding, and named pipes. Today corporations large and small are deciding which operating system they should use for which applications in their organizations | excel, dataset | source | 5 |
| 193 | Taking Advantage of Xerox 3700/4050 Printing Ability within a SAS Batch Processing Report | David Yui | This paper will describe how SAS" is used together with the XEROX printer codes to generate the appropriate combinations of documents to be printed on the XEROX 3700/4050 printer. The | data step, macro put, dataset | source | 6 |
| 194 | Techniques for Using Screen Control Language in an FSEDIT Multi-User Environment | Ann E. Carpenter | SAS/SHARP and SAS/FSP~ software work together to allow conv current update access to SAS" files. Screen Control Language (SCL) applications written with the FSEOtT procedure in a multi-user environment must be aware of locking restrictions imposed by SAS/SHARE software to preserve data integrity. This paper dis cusses techniques for using SCL in FSEDIT multi-user applications. You should already be familiar with the FSEDIT procedure and Screen Control Language |
where statement, libname statement | source | 4 |
| 195 | Testing the Rationality of Panel Data Forecasts Using a GMM Estimator | Richard H. Cohen | residual, correlation, covariance, sample, panel data, time series analysis, null hypothesis, standard deviation, binary variable, regression analysis | source | 6 | |
| 196 | Tetraplot - Four Vertices Are Better Than Three | Leon L. Fedenczuk, Mark Bercov |
Last year we gave you the T emary Plot - this year we have added another vertex and present the Tetrahedron Diagram. This diagram allows the user to display four coorcfinate system data in a three dimensional space. Each apex of the tetrahedron represents a single parameter or a scientifically valid combination of parameters in the original data under investigation. In addition, we have increased the dimensionality of the system by providing for user selected shapes and colors. The Tetrahedron Diagram is a powerful extension of the previously presented Ternary Plot. It allows for the classification of data based on the spatial coherence of data subsets. As a consequence, this graphical tool increases the ability of the user to easily display and interpret information for complex systems such as: chemical analysis, composites of rocks and various multiphase systems, or the output from multivariate statistical procedures | proc princomp, numeric variable | source | 5 |
| 197 | The Art of Defensive Coding | Eva Quint | proc step, macro variable, audit trail, file statement, macro facility, proc contents | source | 8 | |
| 198 | The Colour Centrifuge! | Philippe Brown | data transformation, bar chart | source | 3 | |
| 199 | The Consortium Alternative: Getting the Best Training for Less | Michael L. Davis, Peter Prause |
In June, 1990, The Hartford Area SAS Users Group (HASUG) realized there was little SAS training offered at any of their member companies. As a result, HASUG began an aggressive program to sponsor SAS Institute consortium courses. As of April 1992, these courses saved member companies approximately $250,000 while training about 360 students. Along the way, HASUG learned about the requirements of supporting consortium classes. This included the need to publicize courses and to develop a sense of shared ownership. We also learned that successful consortium courses require attention to a myriad of details. Last, we found that consortium education can be an important function for local SAS user groups. This paper gives an overview of all SAS training alternatives and provide a flowchart to help guide you to the appropriate training solution. This paper describes techniques that may be used to run courses more smoothly | sas programming, macro facility | source | 5 |
| 200 | The Development of a Flasher Tracking System | Linda Ivey | dataset, group by | source | 6 | |
| 201 | The DynaLink System | David Sommer, David Goossen, Andrew M. Traldi |
data dictionary, sample, proc step | source | 6 | |
| 202 | The Estimation of Drug Shelf-Life | Jyh-Shiun Lin, Shein-Chung Chow |
In the pharmaceutical industry, a sta bility study is usually conducted to es tablish the drug expiration dating period(or shelf-life) of a drug product. The method suggested in the current FDA guidelines does not consider the situa tion when there is a batch-to-batch variation. Chow and Shao (1991) proposed a model which can take into account the batch-to-batch variation | sample, macro variable | source | 6 |
| 203 | The Evolving Role of the SAS System in the Development of the Production System for the Consumer Price Index (CPI) | James E. Johnstone, Stuart Leven |
The Bureau of Labor Statistics (BLS) produces a number of economic indicators measuring changes in various market sectors, among them the Consumer Price Index (CPl). This index measures, using a probability sample. the change in price of a fixed market basket of consumer goods and services in specific geographic areas as well as by region and across the U.S. The impact of the index is far reaching. In the 1986 SUGI Proceedings. the role of the SAS System in the development of the revision of the associated production system was discussed. This revision came on-line in 1987. This paper discusses experiences in five years of production use of the SAS system in this environment, the faults and foibles uncovered. and the role of the SAS system in maintenance in a real-life setting of this large-scale system. It also addresses the role of the SAS system as foreseen in the next revision of the CPl. scheduled to come on-line in 1997 | sample, macro facility, imputation, macro variable, proc step, macro definition, dataset, proc sql | source | 10 |
| 204 | The GREPLAY Procedure: A Valuable Tool for the SAS Graphics User | Kimberly A. Mitchell, C. Michael Dunham, David R. Gens |
PROC GREPLAY is a multipurpose presentation tool for creating images from information generated by SAS graphics procedures. Greplay allows the user to view and modify gmphics catalogs which contain the image producing details. The user can inodify the names and colors of graphs, select specific graphs for viewing, move graphs between catalogs, delete graphs, display more than one image per screen, and transmit image producing information to different drivers. Once the catalog is created, Greplay eliminates the need for repetitious creation of data steps and graphics procedures when multiple image visualization is necessary. The user has the option of using interactive or batch mode environment. PROC GREPLAY has proven to be a valuable technique for storing, modifying, and presenting SAS generated graphics display information |
proc greplay, proc step | source | 5 |
| 205 | The History of the SAS System at the Census Bureau | Michael Fortier | statistical software, proc freq | source | 3 | |
| 206 | The Investigation of Suicide Using SAS Software Procedures: A SAS/ETS Approach | Seid M. Zekavat | anova, correlation, predictor variable, standard deviation, estimate | source | 6 | |
| 207 | The Mighty Macro versus the Lowly Link | Barbara M. Ison, Dania P. Ferguson |
With the advent of macros and the resulting publicity and interest in their use, the power and efficiency of the LINK may easily be overlooked for many appropriate applications. Efficiency needs to be a concern to us all, particularly those of us using mainframe time'sharing environments. The CPU/cost savings potential of techniques such as those described below are substantial, especially when developing SAS' systems to be used repetitively in a production mode. The efficiencies we found under MVS hold true for other operating systems running SAS. This paper will describe when LINKs can, and probably should. be used instead of macros. We will present our real·life experiences as examples to demonstrate the power. similarities and differences in these two approaches as well as the relative efficiencies | macro definition, macro facility | source | 6 |
| 208 | The Parke-Davis Assessment Program - An Example Using SQL in SAS Programming | Russell Newhouse | study day, data step, proc sql, proc step, create table, group by, baseline | source | 6 | |
| 209 | The REPORT Procedure: An Introduction to the Batch Language | Ray Pass | proc report, proc step, sample, proc print, class variable | source | 10 | |
| 210 | The SAS System Enters the Third Dimension Using DATA Step Views for Real-time Quality Control | Jeffrey A. Polzin, Robert N. Rodriguez, James F. Sattler |
The SASsSystem is developing rapidly and in many ways as it approaches the 21st century. This paper deals with one set of quan titative changes and shows how these changes, in combination, usher in a new qualitative change. As a result of new developments at the supervisor, procedure, and host levels, the SAS System is moving beyond batch and interactive processing capabilities and into a new, third dimension of real-time and event-driven processing. This opens up entirely new vistas for applications developers | sample, sysrc, dataset, proc step, real-time processing | source | 5 |
| 211 | The SAS System Supervisor - A Version 6 Update | Merry G. Rabb, Donald J. Henderson, Jeffrey A. Polzin |
This tutorial updates a talk given at previous SUGls that discussed the functions of the SAse System supervisor during the compilation and execution of a SAS DATA STEP program. Under Version 6 of the SAS System. the SAS System supervisor provides interface ser vices at a higher level than before, and its primary activity in this context is to detennine whether a DATA or PROe step is being invoked. Once invoked, the OAT A step controls its own environ ment during both compile and execution phases. This tutorial pres ents a logical mOdel describing both the compile and execution time actions of the OAT A step processor |
by group processing, dataset, set statement, update statement, keep statement, merge statement, missing value, where statement, sample, conditional do loop, drop statement, output, sas programmer, index, proc step | source | 8 |
| 212 | The SQL Procedure as a Tool for Natural Language Processing | Kurt Schmidt | A SASe based system for translation of natu ral language data into standardized forms or codes more readily analyzed by computerized means is presented. In survey research and other areas information is frequently collected as natural language utterances. These either cannot be processed by current data pro cessing tools or have to depend on complex and highly specialized tools. This leads to potential losses of data or high processing costs, as either the questions have to be very limiting, have to be painstakingly translated into uniform form by humans or information has to be foregone. With the advent of SASe SOL Procedure and SASS Macros across virtually every platform, it is now possible to implement a SAS based system that performs translation of large percentage of such free form natural language data into standardized form readily manipulated by other computer ized tools, a task previously thought of as an exclusive domain of specialized artificial intelligence tools | proc freq, data wrangling | source | 6 |
| 213 | The Stand-up Trainer Is Not Yet Obsolete | Steven J. First | source | 3 | ||
| 214 | The Use of Logit Analysis: Pitfalls and SAS Tips | William Chan | The log it model is being used with increased frequency by social scientists -- sometimes indiscriminately. In this paper we address some basic assumptions of the logit model and highlight some common pitfalls in the applications of the logit model. A few SAS@ tips which could correct some of the technical problems are also discussed. An example of a study of professional affiliation in a medical society is used to illustrate the non-technical interpretation of the log it model. During the past decade, the logit model became a popular choice as a multivariate technique for analyzing multidimensional contingency tables where both dependent and independent variables are categorical rather than continuous. Approximately 54 percent of the articles containing contingency tables published in the American Sociological Review in 1989 used this model. Despite its popularity, technical flaws are found in many articles | logistic regression, response variable, predictor variable, proc step, interaction effect, sample, proc freq, main effect, regression analysis | source | 3 |
| 215 | The Use of SAS/GRAPH Software in the Management of Clinical Trials | MaryAnne DePesquo Hope | At the Veterans Affairs Cooperative Studies Program Coordinating Center graphics are important tools in monitoring and presenting data from clinical trials. SAS/GRAPH' is used to display a large amount of longitudinal data covering all phases of the trial from the first protocol to the final publications. Customized reports are created with the Annotate facility and the procedures available through SAS· and SAS/GRAPH software. A wide variety of graphic techniques and applications including line graphs, bar charts, and whisker plots are used to summarize and present data from the clinical trial on patient intake, drug compliance, primary endpoints and secondary endpoints. In addition, maps and pie charts are used to display data for administrative purposes including cost factors and participating sites for various studies |
proc step, clinical trial, secondary endpoint, proc gchart, line chart, variable, macro facility, proc means, proc gplot | source | 6 |
| 216 | The Utter Simplicity of the TABULATE Procedure | Daniel E. Bruns | proc step, proc tabulate | source | 5 | |
| 217 | The View Review: PROC SQL Views, SAS/ACCESS Views, and DATA Step Views | Andrew A. Norton | proc sql, where statement, proc step, dataset, view | source | 6 | |
| 218 | Training and Supporting Nontechnical People Who Use theLAS System | Kim L. Kolbe | Frequently, the people who use the SAS System the most are those individuals who are very accomplished in their respective field, but are far from being full-fledged programmers. This makes the job difficult for the training and user support personnel because often times these users know precisely what they want, but cannot get past the technical details to produce the desired results | source | 3 | |
| 219 | Transporting Files between Host Operating Systems | Yvonne Selby | With Release 6.06 of the SAS" System and enhancements provided in Release 6.07, transporting SAS files between various host oper ating systems is becoming more common. This article discusses the SAS data library engines and procedures used to move SAS data sets. catalogs, and libraries between releases of the SAS System and host operating systems. Enhancements to the CPORT proce dure are also discussed | proc step, catalog entry, proc cport, proc copy, proc cimport, catalog, libname statement, dataset | source | 5 |
| 220 | Twin Studies: Now That I Have Your Attention, or How One Users Group Kept Its Crowd After Lunch | Lynne A. Davidson | The Virginia SAS Users Group holds quarterly meet ings to promote SASS software usage and proficiency. The standard meeting format includes a business meeting and one or two speakers in the morning, a lunch break, and then small group sessions and/or a panel discussion after lunch. In order to deal with the tendency for members to leave before the after-lunch sessions, the program oommittee attempted an innova tive panel presentation. This involved a problem presentation and several solution suggestions presented by Users Group members. The general interest of the topiC and the applied solutions proved effective in maintaining a higher than average attendance in the afternoon session | proc step, proc sort, proc sql, create table, proc transpose, sample | source | 5 |
| 221 | Use of SAS/AF and SAS/FSP Software to Integrate a Payment System for Patient Information in a Large Multi-Center Clinical Trial | Dorothea Collins, Margaret Lee |
The data management in a multi-center clinical trial often involves the tracking of a large quantity of patient information that accumulates while the intake phase is progressing. In addition, hospitals participating in the study are provided with the necessary support to perform randomization and follow-up procedures. In recent large cooperative trials run in the V A, this entails reimbursement to hospitals based upon the number of patients randomized and the number of follow-up forms received at the central coordinating center. SAS/AF, SAS/FSP, and SCL were used to design a system to manage these accounting functions and integrate this information with the extensive data management system for the study written in SAS@ System Version 6 for mainframe and microcomputer processing. This payment system is expected to increase the receipt of data as well as the quality of data which will be checked before payment |
randomization, clinical trial, proc step, data governance | source | 6 |
| 222 | Use of the KEEP and DROP Concept in Batch SAS Programs | Bill Schallert | keep statement, dataset, set statement, proc step, input statement, proc sort | source | 6 | |
| 223 | Using and Understanding SAS Formats | Howard Levine | The purpose of this paper is to enable you to use SAS formats to perform the following tasks more effectively: Improving the Appearance of Reports Reducing storage requirements for SAS data sets Using formats to perform table look ups [Relational Data Base] Using informats to read data Converting SAS Formats to SAS data sets and vice versa Permanently storing SAS formats Topics In order to meet the objectives, these general topics will be expanded upon: Types of SAS formats Syntax for creating simple formats Examples of how to use formats Tvpes of SAS Formats There are three ways that a format can be classified. Every format will have one characteristic from each of the following groups. 274 Character or Numeric Just as all SAS data set variables must be either character or numeric, so must SAS formats and in formats be character or numeric. Character formats have their names start with a II $" while numeric formats do not | proc step, proc format, format statement, dataset, char., proc print, put statement | source | 7 |
| 224 | Using Features of the SAS System, Release 6.06, to Enhance Data Entry Applications | Sharon S. Trevoy | data cleaning, data integrity | source | 4 | |
| 225 | Using Macros to Run Regressions | Reesa L. Laws | Anytime a programmer has to repeat code, macros should come to mind. Macros can cut programming time and code by half. This paper describes how macros are used to run regressions and to create intercept and prediction variables in any model a user chooses. The user needs only to select the outcome variable, the model and give the number of variables in that model. With the macro described, the programmer need not change the code when a different model of varying length is selected | macro variable, proc reg, proc step, regression analysis, data cleaning, response variable | source | 2 |
| 226 | Using Personal Computers as Host Graphics Devices | Mike Kalt, Peter Ruzsa, Mike Long |
This paper discusses numerous alternatives for using personal computers to display, save, and produce hardcopy of graphs created on remote hosts. Topics discussed include terminal emulation, cooperative processing, and transferring graphics catalogs and files | proc step, catalog entry, proc greplay | source | 6 |
| 227 | Using Pipes to Help Your Plumbing | Greg Silva | Searching for system information while running a SAS' program can be difficult under the easiest of operating systems. When you want some informa tion while running under Unix2 , then problems grow. With the help of the pipe argument in the FILENAME statement, however, you can easily 'plumb the depths' of your system from inside of a program |
filename statement, macro variable | source | 3 |
| 228 | Using SAS Econometric Modeling to Predict Year 2000 Census Mail Response | Jay K. Keller, Michele Carrier |
regression analysis, response variable | source | 4 | |
| 229 | Using SAS Macros and SAS/GRAPH Software to Produce Data-Driven Laboratory Data Plots | Paul M. Gilbert | Data driven progrcmuning is a method of programming where the values of the data ftdrive n or create the SAS code. This technique is illustrated using SAS macro to read a dataset and create code to produce customized reports and plots. The data driven macro produces SAS code to subset data, sort data, create annotate dalasets, create PROC statements, create formats, create titles and create footnotes. This code is created and executed for each similar subgroup of data contained in the dataset. Source code is provided |
proc step, in=, clinical trial, laboratory data, macro variable, macro definition, %if-%then-%else, macro loop | source | 5 |
| 230 | Using SAS Macros to Generate Code (So Your Program Can Figure Out What to Do for Itself) | Nancy K. Blood | Ever find yourself trying to maintain multiple versions of essentially the same program? Each version may produce summary statistics for different variables (for example, HEIGHT and WEIGHT), or select different subsets of data for analysis (males, females, or age groups) but the numeric variables, the labels and format statements remain the same. The techniques presented here use the SAS macro facility to demonstrate how designing your programs to be adaptable eliminates the need for multiple versions. You will structure your programs to use unique parameters you identify to generate the desired code | proc step, %if-%then-%else, macro definition, proc means | source | 5 |
| 231 | Using SAS Software to Deliver an Offender Population Profile System (OPPS) | Larry Motiuk, Roger Boe |
dataset, data integration | source | 5 | |
| 232 | Using SAS Software to Develop a Geographical Database for Aerial Photography | Shlomit Yaari | A database containing information on more than 20,000 aerial photographs was established at the Department of Geography and the Computer Center of Bar-Han University. Aerial photography provides a unique source of information for mapping and landscape changes. The database was constructed using SAS as a database management system, integrating the following products: SAS/BASE, SAS/AF, SAS/FSP, SAS SQL/DS and SAS/GRAPH. A full-screen application was developed using the menu structure of SAS/AF software which enables the researcher to update, retrieve, summarize and display data. The data collected is maintained by IBM' s VM relational database SQL/DS and can be extracted to create a SAS data set, using the SAS-SQL/DS interface. The data set created, containing records meeting certain criteria defined by the user, can be further utilized for a variety of statistical analyses. SAS/FSP is used to add to and update the data, which is automatically checked for inconsistencies and accuracy |
proc sql, dataset, macro definition | source | 6 |
| 233 | Using SAS Software to Develop a User-Friendly Application Interface | Elizabeth Ogden, Kevin Steiger |
This paper discusses how we used SASjAF, SAS Screen Control Language (SCL) , and SAS Structured Query Language (SOL) products available in SAS version 6.06 software to develop a user-friendly database application. It illustrates how we used short and long selection lists to implement user-interface panels that display all of the correct entries for a field, automatically validate user-input, and supply error messages. For the case where a user Input field contained too many possibilities to be listed, we describe a method to combine selections from a series of lists. Lastly, we show how we used extended tables for a database application where the user needs to specify particular SAS dataset observations | catalog entry, error message, proc sql | source | 5 |
| 234 | Using SAS Software to Evaluate the Imaging Performance of Presurgical Diagnostic Tests in the Detection of Cancer | Kristin R. Dawson | Several presurgical imaging modalities such as CT scan, chest x-ray. MRI, ultrasound, and bone scan, have been used for detecting the presence of cancer in humans. CYTOGEN Corporation has developed a radiolabeled monoclonal antibody to be utilized as another type of imaging modality for the detection of cancer. The process of evaluating the ability of diagnostic tests, including immunoscintigraphy (antibody imaging). for detecting disease in patients involves comparison of the results of these tests to the surgical findings. The procedures used to evaluate imaging performance parameters, such as sensitivity and speCificity. will be discussed in this paper, and the uses of SASS software to analyze the data are presented | type i error, type ii error, clinical trial, do over, subgroup analysis, macro definition | source | 7 |
| 235 | Using SAS Software to Evaluate the Power of Alternative Data Analytic Methods for Repeated Measures Designs | Thomas C. Chenier | sample, repeated measures, sample size, covariance, effect size, proc univariate, covariance structure, proc step | source | 6 | |
| 236 | Using SAS Software to Teach a University Course in Statistical Computing and Database Management | John L. Wasik | proc step, data science, sas programming, data governance | source | 6 | |
| 237 | Using SAS Software Windowing Products to Develop a Remote Data Entry Diskette Tracking System | John Bowen, Cathy Carfagno |
where statement, macro variable, dataset | source | 6 | |
| 238 | Using SAS/Access Software and Database Management Systems in a Distributed Environment | Jana Van Wyk | The growth of distributed computing environments that use more than one database management system (DBMS) and more than one operating system has made interoperabiJity among data, software and hardware resources increasingly important. Using features of the SAS System as tools to access and analyze these data across the network can make this interoperability possible. There are three types of software that can be used to build a distributed application. First, SAS/ACCESS software allows SAS users to access data in numerous database management systems on a variety of hardware platforms. Second, many database management systems offer transparent distributed access support for remote databases. And, third, SAS/CONNECTTM software offers SAS users connectivity between SAS sessions and numerous operating systems and hardware configurations. This paper describes a distributed SAS application that utilizes these three types of software in a distributed environment |
oracle, select, proc step, proc download, dataset, oracle access, proc sql, proc print, bar chart, data science, data step | source | 10 |
| 239 | Using SAS/ACCESS Software for Rdb in Clinical Trials Research | Joseph B. McGlynn | In the past, clinical trials data were stored in an assortment of separate datasets. Each new clinical program required new datasets and new reporting and analysis programs. In an attempt to reduce the amount of programmer support required for clinical trials, a relational database was developed using Digital Equipment Corporation's Rdb. The goal was to provide a stable database that would capture a minimum of 80% of all clinical data collected. All rules governing the management and usage of the data were built into the database instead of the applications programs. We hoped to reduce errors in programming and speed data analysis as a by-product of this effort. The tools used in the development of this project include V6.06 of the SAS System including SAS/ACCESS, SOL, Rdb/VMS, and VAX Rally. This paper presents the results of this effort |
clinical trial, relational database | source | 4 |
| 240 | Using SAS/AF Software and SCL in Version 6 | Karen Crandall | inputn, missing value | source | 4 | |
| 241 | Using SAS/CALC Software to Complement SAS/STAT Software | Edward Wei-Tarng Lee, Kendall Bryant, Mary-Ellen Power |
The ANCOVA (Analysis of Covariance) and KAPPA statistics are widely used in epidemiology and clinical trials research. The purpose of this paper is to describe two applications of the spreadsheet software to automatically compute: 1) the interaction point and 95% Confidence Interval when the parallelism assumption is violated in ANCOVA; 2) KAPPA statistics. In addition, a comparison of functions of SASjCALC for WindowslM , Lotus 123 for OSj2.fM v.1.1, EXCEL 4.0, Quattro Pro 4.0 and an introduction to using spreadsheet software for data base management and graphic presentation |
dataset, excel, confidence interval, cohen's kappa, proc step | source | 10 |
| 242 | Using SAS/GRAPH Software for Three-Dimensional Illustrations of VL Amino Acid Diversity7 .. | Perry Watts, Samuel Litwin |
Antibodies have a pattern of amino acid sequence diversity that is intrinsically related to their physical structure. Highly specialized equipment such as the Evans & Sutherland PS390 graphics system provides a realistic three dimensional display of an antibody's backbone. Amino acid residues are represented as colored spheres where hue is an indicator of diversity. However, the system is expensive, and the color displayed is influenced by a point's depth coordinate. 1bis means that there is a discrepancy between an explanatory legend positioned in the foreground and the antibody that is being rendered. Reproduction of hard-copy results for journal publication requires photography instead of the more convenient color laser printing. Given these problems, there is a role for PROC G3D. This procedure is used to draw a molecular disc and stick model with hues representing varying degrees of diversity among the residues |
proc step, lag function | source | 6 |
| 243 | Using Task Broker to Enhance SAS Batch Processing in a UNIX Environment | Lori Goss | In a production environment, it is quite often necessary to have numerous SAS R jobs competing simultaneously for computing resources. On a UNIXR network it is possible for the SAS Programmer to selectively request these resources. However, it is highly probable that optimum specifications on one day will be less than optimal on another day. It becomes an overwhelming task for a Programmer to monitor the computing resource fluctuations. Therefore, the need arises to dynamically select resources based upon the current load on computers on the network at any given time. Hewlett-Packard (hereafter known as HP) has developed a product, Task Broker, to aid in resolving this dilemma. Task Broker distributes a submitted task to the most capable computer on a network. This computer is selected based upon predetermined criteria such as its ability to perform the task (i.e. accessibility to software and data), its current load average and the task priority | sas batch, batch job, options statement, sample | source | 7 |
| 244 | Using TeX to Format SAS DATA Step Reports | Joseph E. St. Sauver | sample, character functions | source | ||
| 245 | Using the ARIMA Procedure in Forecasting the Demand and Utilization of Inpatient Hospital Services | John J. Hisnanick | moving average, proc arima, estimate, seasonality, stationarity, proc step, irregular component | source | 9 | |
| 246 | Using the CATMOD Procedure to Estimate Linkage between Pairs of Gene Loci from Offspring of Selfed Heterozygotes | Kent M. Eskridge, William Compton, Lomas Tulsieram |
Estimation of linkage between pairs of gene loci is an important problem in genetics. The strength of linkage between pairs of gene loci is measured with the recombination value r and is called the genetic map length. The recombination value is the probability that a gamete of an individual has resulted from a crossing over of genes between the two parental gametes received by that individual. When each of two gene loci has two genes (A,a and B,b), four gametes are possible (AB, Ab, aB, ab). The calculated number of individuals in each genetic class from the phenotypes of selfed double heterozygotes may be used to estimate the recombination value. Computer programs are available that give maximum likelihood estimates of r. However, homogeneity tests of r values are not easily performed, nor are these programs conducive to processing large data sets | likelihood function, maximum likelihood estimation, generalized least squares, proc step | source | 6 |
| 247 | Using the FSLETTER Procedure in the SAS Macro Facility to Produce Quality Letter Output for a Marketing Department | Phyllis Jordan-Johnson, Margaret L. Molnar |
proc step, index | source | 9 | |
| 248 | Using the GLM Procedure to Analyze Multivariate Response in Crossover Designs | Julie Myers Grender, William D. Johnson |
The crossover design involves observing the same response variate under different experimental conditions for each subject, and as such, is a special case of a repeated measures or multivariate data structure. The general multivariate linear model provides a natural framework for analysis of such data because generalizations to more complex structures are feasible and follow the same rationale. Response to treatment is often multivariate, especially in clinical studies, although multivariate response has received relatively little attention in the literature on analysis of crossover designs. The GLM procedure provides an easy-to-use approach to the analysis of multivariate response data arising in crossover stodies. We demonstrate the methods and outline the steps for using SAS· to carry out a general class of analyses |
treatment effect, treatment period, proc glm, proc genmod, repeated measures, covariance, longitudinal data | source | 6 |
| 249 | Using the Make Routine to Aid System Development | Greg Silva | The UNIX' operating system comes with a weaHh of tools. Among them are routines that can be used to aid system development. The purpose of this paper is to show how one of these tools, the make utility, can be used to help the system developer automate some of his or her tasks | macro definition, dataset | source | 4 |
| 250 | Using the NLIN Procedure to Combine Estimates of Relative Potency | Stan Altan, Jaya Natarajan |
The estimation of relative potency from indirect biological assays, assuming parallel line model has been thoroughly discussed by Finney(1978). Finney has also given a method for combining estimates of relative potency when the same test preparation is evatuated over s (s> 1) assays. Finney's formula is not applicable when pooling estimates from k (k> 1 ) replicates of the same tesl preparation from a single assay. In this case, the replicates are correlated due to a common standard. The current paper discusses two methods for finding the appropriate weights for the pooled estimate of relative potency in the case of correlated estimates. The first method uses the formulae for approximate standard errors of functions of random variables (Stuart & Ord, 1987) to obtain the covariance matrix, while the second method looks at a non-linear reparameterization of the model. A numerical example is also given using the NUN procedure to produce all necessary parameter estimates | covariance, regression analysis, proc iml, estimate, proc nlin, proc glm | source | 6 |
| 251 | Using the Project Management Tools in the SAS System | Radhika V. Kulkarni, Gehan Corea |
The project management procedures in SASjOR@ software have been growing in scope con tinuously over the years. The first part of this paper briefly lists some of the new options that are available in the CPM, GANTT, and NETDRAW procedures. The major focus of the paper, however, is to describe and illustrate some of the more advanced scheduling features of the CPM pro~edure. The CPM procedure allows you to specify target dates, save baseline dates, update a project that is in progress, schedule subject to scarce resources, and perform several scheduling tasks. Each of these features of the procedure can be fine tuned according to your needs using the options available. To effectively use any project management system, it is essential to understand the heart of the system; in particular} you need to understand how all the scheduling options work together. New Features Several new options have been added in Release 6.07 (SAS Institute Inc., 1992) |
source | 13 | |
| 252 | Using the SAS BMDP and SPSS Engines with Release 6.07 of the SAS System and VMS 5.4 | Barbara B. Okerson | This paper discusses the function and usage of the BMDP and SPSS engines under version 6.07 of SAS using the VMS operating system. The paper begins with an overview of the SAS Multiple Engine Architecture and the differences between interlace and native engines. This is followed by detailed looks at the SPSS and BMDP engines with examples for read-only access to SPSS and BMDP system files, and conversion to permanent SAS data sets. All code used is included | dataset, proc step | source | 3 |
| 253 | Using the SAS System for an Analysis and Reporting System for Effluent Monitoring | Thomas P. Walker Jr. | A computerized analysis and reporting system was developed for personnel involved in effluent monitoring operations at the Oak Ridge Y -12 Plant managed by Martin Marietta Energy Systems. lnc., for the United States Department of Energy under contract DE-AC05- 840R21400. SAS/ACCESS· interface to RdbNMS·, base SAS software, SAS/GRAPH·, SAS· BATCH, the macro facility, and some aspects of Digital Command Language (Del 1ll ) were used in the development of the reporting and analysis system. This system essentially contains three modules, each of which has its own function. The basic functions include the creation of monthly reports for easy perusal for excursions; run charts of the observational laboratory data; and a non parametric trend analysis of monthly summary values. which involves the computation of the Mann-Kendall test for slope |
dataset, trend analysis, macro facility | source | 6 |
| 254 | Using the SAS System for Medical Data Processing and Event Analysis: An Overview | Harald Pitz, Marcus Frenz, Hans-Peter Howaldt |
This paper shows necessary methods and appropriate SAS utilities (data and procedure steps) to perform medical data analysis. The | proc step, proc format, data step, proc phreg, interval variable, data science, data wrangling, dataset, date variable, proc freq | source | 6 |
| 255 | Using the SAS System to Evaluate Syphilis Control | Raymond L. Ransom, Joann Schulte, Melinda Flock |
dataset, proc download | source | 2 | |
| 256 | Using the SAS System to Supervise Disk Pack Usage in an MVS Environment | Peter Yribar, Michael Sternberg |
The SAS system is widely recognized as an easy to learn end. user language, as well as a powerful statistical tool. This paper will focus on another capability of SAS, as useful tool for systems applications. Specifically the utilization of an MVS volume table of contents (VTOC) to supervise disk pack usage for an entire deparunent. This is possible in base SAS, using the VTOC option on an INFILE statement. With this capability in combination with the SAS macro language it is possible to report on all facets of disk pack usage, as well as delete datasets that do not meet specific predefmed guidelines. In short we are able, in one program to maximize disk utilization | macro variable, macro definition | source | 4 |
| 257 | Using the SAS System within a Strategic Information Architecture | Tim Feetham | proc freq, proc step | source | 4 | |
| 258 | Using the SQL Procedure | Kirk Paul Lafler | 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 SELECT statement, and perform efficient queries against SAS System data sets. Prerequisites This workshop introduces the syntax and uses of the SQL Procedure. No prior knowledge of SQL is required since this material is designed to acquaint the , user and/or programmer to the many features found within the SQL Procedure |
proc sql, proc step, group by, order by, select, create table, relational database, update statement, where statement | source | 6 |
| 259 | Validation of the UPLOAD and DOWNLOAD Procedures in the SAS System | Richard A. Markus, Jeffrey S. Kouba, Curt Snyder, Louise Murphy |
SAS provides the Upload and Download procedures for transferring data sets between a microcomputer and a host. The integrity of transferred data is critical. Ideally, the transferred data sets would be identical to the original data sets. Unfortunately, this is not likely since there are differences in machine precision. Different operating systems store values in different formats, such as EBCDIC and ASCII. This validation examines the magnitude and implications of the differences created by transferring SAS data sets between a microcomputer and a host | clinical data, raw data, proc step, data cleaning | source | 6 |
| 260 | Weather Sensitive Appliance Load Curves: Conditional Demand Estimation | Deepak A. Keshani, Thomas N. Taylor |
regression analysis, time series analysis | source | 9 | |
| 261 | What's New in Release 6.07 of SAS/GRAPH Software | Donna Bravo | The request for a graphics editor has been a top SASware Ballot" vote-getter for many years and, now, Release 6.07 of SASlGRAPH software brings you the ability to edit graphs interactively within the SAS" System. With the graphics editor, you can bu~d your pictures from scratch or easily enhance your SAS/GRAPH output with your own imagination or other graphs. Enhancements to graphics procedures in Release 6.07 include the ability to label contour lines automatically and to control label and contour line attributes; to print the regression equation used in GPLOT output; to display a legend for overlayed plots; to produce error bars for horizontal and vertical bar charts and to route graphics output automatically via new GDEVICE procedure screens |
proc step, proc gchart, regression analysis | source | 9 |
| 262 | X Window Applications and the SAS System: A Partnership in Quality | Susan G. Smith | The perception of quality service is often related to the timely and accurate response to user concerns. To help us assess how we are doing, we partnered an X window application with SAS to input and analyze data about user contacts. During the project, we ran into a problem with null bytes in C-generated binary files. This paper describes our solution to the handling of raw C binary files and how we hope to increase our quality of service by combining the use of an X application with SAS | sysdate9, date variable | source | 3 |
| 263 | YTD Reporting Using the SUMMARY and TRANSPOSE Procedures | Tom Bellmer, Steven J. First |
This paper reveals techniques using PROC |
proc step, data step, proc means, missing value, macro variable, proc transpose | 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.