HOME
QUICK SEARCH
AUTHORS
KEYWORDS
HELP
BUGSSUGI 1984 Papers - Index
180 papers. Source: SAS Users Group International conference proceedings.
| # | Title× | Authors× | Abstract× | Keywords× | Extras | Pages |
|---|---|---|---|---|---|---|
| 1 | %Macro for Kaplan-Meier Survival Curve Estimation | Frank E. Harrell Jr., Michael J. Helms, Lawrence H. Muhlbaier |
proc calis, class variable, macro definition | source | 5 | |
| 2 | %Macro UPONE for Capitalizing the First Letter of Each Word | Lawrence H. Muhlbaier, Judith Stafford |
data wrangling, do over | source | 2 | |
| 3 | A Botanical Garden Data Base Implemented in SAS Software | Edward A. Greenberg, Bruce R. Lewis |
This paper describes a data base devel oped for the Phoenix Desert Botanical Garden's Living Plant Collection usin~ Statistical Analysis System (SASR) software. The hierarchical data struc ture is implemented as a relational data base. A batch SAS program is used to apply updates to the data base and pro duce rollback and backup copies on tape. To facilitate data entry fnd changes to the data base, a WYLBUR exec is em ployed as a front-end to SAS. A data dictionary is used to produce data collection forms, drive the WYLBUR data entry program, and produce SAS state ments that are used in the SAS update program. The data dictionary provides a convenient means of achieving consis tency between the contents of the data base and the programs used in its maintenance | data dictionary, sas batch, format statement | source | 6 |
| 4 | A General Approach to Decision Support Systems Utilizing the SAS System | William Ingram III, Thomas P. Rothrock |
data science, dataset, data governance, proc step, proc sql | source | 7 | |
| 5 | A General Woolf-Haldane/Mantel-Haenszel Procedure for Dichotomous or Polychotomous Exposure Variables and for Any Number of Stratification/Control Variables | Philip N. Gallagher Jr., William M. Haenszel |
confidence interval, odds ratio, proc freq, logistic distribution, regression analysis | source | 3 | |
| 6 | A Generalized Interactive Full Screen Edit-Update System Using SAS Software | Jesse Gary, Donald J. Henderson, Connie G. Welch |
proc step, audit trail, sample, master data management, dataset, data step, merge statement, transaction data | source | 6 | |
| 7 | A Hierarchical Approach to Daily Stock Market Forecasting | Heydar Pourian | proc step, residual, regression analysis, proc arima, proc reg, autocorrelation, correlation | source | 6 | |
| 8 | A Human Resource Information System Using SAS Software | Alan A. Parrow | proc step, data step | source | 6 | |
| 9 | A Macro for Fitting Nonlinear Models to Poisson Distributed Data | Edward L. Frome, Richard W. McLain |
Models are considered in which the depen dent variable y 1s a count that follows the Poisson distribution. The expected value of y Is represented with a regression function that describes the relation between the expeeted count, the predictor variables, and the parameters. Estimates of the parameters can be obtained using iteratively weighted least squares (IWLS). The IWLS procedure is equiva lent to using the method of scoring to obtain a root of the likelihood equations. Poisson re gression models include linear, log-linear, quasi-linear, and intrinsically nonlinear re gression functions. This approach allows the analyst to concentrate on determining the most appropriate form of the regression function without regard for computational complexity | regression analysis, poisson distribution, poisson regression, likelihood function, maximum likelihood estimation, proc nlin, estimate, proc step | source | 6 |
| 10 | A Phosphate Product Mix Model Using the LP Procedure | Agnes Eggleston | proc step, proc transpose | source | 5 | |
| 11 | A Procedure for Selecting Multiple-Item Cutoff Samples | Eugene M. Burns | sample, sample size | source | 6 | |
| 12 | A Queueing Theory Model for Optimization Using SAS Software | Chitra B. Prabhakar | This paper describes a study undertaken to determine the optimum number of MVS computer terminals required in a computer center to achieve the greatest productivity for the lowest cost. A queuing theory roodel was used, where the prCXjrammers are considered as customers and the terminals as servers. Distributions of inter-arrival times and service times were estimated using sample data. A large number of arrivals and services were simulated using the estimated service time and inter arrival time distributions. Under suitable assumptions, estimates of the waiting cost were obtained from these simulations, for different number of terminals in the system. From these measures, the optimum number of terminals and various parameters of interest for the corresponding queue were obtained. SAS was used for analyzing sample data, for performing simulations of the queueing system and for generation of reports and graphs for the management | sample, monte carlo | source | 4 |
| 13 | A Review of SAS Interfaces with Various Software Systems | Richard W. La Valley | dataset, proc step, relational database | source | 7 | |
| 14 | A SAS Data Base Updating System for Clinical Data | Paul J. Mailloux, George C.S. Yu |
dataset, audit trail, clinical data, proc print | source | 6 | |
| 15 | A SAS Macro for Determining the Causes of Interaction in a Two Way Model | Edward P. Barry, Dar-Shong Hwang |
proc step, inferential statistics, proc genmod, sample | source | 6 | |
| 16 | A SAS Macro for Increasing the Reliability of a Binary Test Using Item Interactions | Robert J. Jannarone, James S. Roberts |
This paper first reviews the concept of "item interactiveuess" and then shows how information from item interactions on a binary test can provide significant, additional information to the researcher. A SAS macro that performs all the necessary tests to assess aud capture this new information is then described. Intr-oduction What are item interactions? This question would be most suitably answered with an example. Suppose a clinical psychologist was attempting to develop a paper and pellcil test to aid in distinguishing between schizophrenic and schizotypal clients. (Both of these conditions may involve oddities of thought, perception, speech and behavior | anova, correlation, sample, data cleaning | source | 5 |
| 17 | A SAS Macro for Performing Repeated Significance Testing in Censored Survival Analyses | Gordon W. Davis, James F. Ward |
A common practice in lengthy clinical trials with serial patient entry is the performance of repeated significance testing to determine if significant | event, hypothesis testing, macro definition, covariance, clinical trial, correlation | source | 6 |
| 18 | A SAS Program for Correspondence Analysis Using PROC MATRIX | Bernard Clement, Daniel Levesque |
proc corresp, proc freq, proc princomp | source | 6 | |
| 19 | A SAS Program for Non parametric Tests for Homogeneity of Dose Response Growth Curves | Saeid B. Amini, Kantilal M. Patel |
A class of nonparametric rank order statistics for testing the homogeneity of dose response growth curves has been proposed and discussed by Koziol et a1. (1981, Biometrics 37, 383-390). These nonparametric tests allow incomplete andlor missing observations on response variable in one or more treatment groups over a period of experimentation. These procedures are applicable when the sta~dard parametric procedures are not appropriate due to missing observations, for example, animal weights in prechronic and chronic bioassays. A general SAS program in PROe MATRIX is developed to compute the Koziol generalized linear rank statistics for testing the homogeneity of dose response growth curves. The program computes the test statistic and the corresponding P-value using i) Wilcoxon rank scores, ii) Van der Waerden normal scores, and iii) median scores | degrees of freedom, null hypothesis, probability distribution, response variable, nonparametric test, sample | source | 6 |
| 20 | A SAS Testscoring and Item Analysis Program Which Utilizes the SAS Macro Facility | Charles E. Rich | raw data, do over | source | 6 | |
| 21 | A SAS-Based System for Producing Tables for a Monthly Circular | Allan Mustard | A system for producing statistical tables, designed for use and maintenance by non-programmers, is presented. The system was programmed and i~ple~ented by agricultural economists to replace parts of a subsystem written in COBOL and an accounting package. It is currently used to create statistical tables used in publication of a monthly circular in the Grain and Feed D'ivision of the Foreig" Agricultural Service, USDA, and si~ilar systems based on it are used in other divisions as well. The system makes use both of TSO command language procedures and of interactive SAS programs to create, modify and submit batch SAS programs. Output is routed variously to paper, to disk, and to remote output queue for transfer to another computer. The SAS-based system is cheaper, more flexible, and more reliable than the system it replaced, and can be modified by analysts with meager data processing and programming skills | proc step, proc printto | source | 4 |
| 22 | A System for the Storage and Retrieval of Citations and the Generation of Bibliographies | Erika Mares Stone | The system stores citation files as SAS data sets which researchers can access through a menu-driven, interactive system under MVS TSO. Citations can be retrieved by searching on authors, titles, subject key words, journal names, or ID numbers. Citations retrieved for bibliographies are written to disk as SCRIPT source files and can be downloaded from the mainframe to a word processor in a form ready to submit to a publisher | dataset, raw data | source | 6 |
| 23 | A Systematic Approach to the Understanding of SAS Macros | Bernie Siryk | A systematic approach to understanding the new macro facili ty is presented. Included in this discussion ~ill be prope~ definition and calling of a macro, iORntifyine e!=lch of thRflR tasks, examining the context in which a macro is called, proper use and passing of macro variables, and a general overview of using macr~ programming statements, Several comEon errors regarding ~acro lnnguagc usage accompanied by a brief | macro variable, macro call, proc step, macro facility, &variable, macro definition | source | 4 |
| 24 | A Text Retrieval System Based on Numerical Encoding | Nelson R. Pardee | Qual itative research frequently generates more data than can be ~sed readily. Researchers using 12,000 pages of text transcribed from ,lOa interviews cannot humanly master the dota In a short period of time. A system of numerically encoding key ideas In the text provides a computer readable; Index for rapid access to areas of Interest in the text | sample, random sampling | source | 4 |
| 25 | ACBPS: A Budget Planning &Reporting System Using the SAS/FSP Product | Jesse Gary, Donald P. Gleason |
dataset, audit trail | source | 6 | |
| 26 | AEPG-Automated Edit Program Generator | Lynn T. Julian | proc step, sample | source | 6 | |
| 27 | An Evaluation of SAS Training Mediums | Kim Edward Trif, Kirk Paul Lafler |
In a fast -paced tecilIlulogica.l indus try, managers are constantly forced to seek compllter training courses for employees and end-users. Traditionally, employers wonder what training courses their employees should attend. But with more training mediums emerging, managers are faced with a new question, which training mediums will best serve the general needs of the company. A true measurement scale for determining the SAS training medium(s) best suited for your needs currently is not available. This paper dis cusses the importance of evaluating and weighing the alternatives and offers helpful suggestions for selecting appropriate training mediums. INTROruCTIDN Todays Data ProceSSing managers are not only concerned with adequate SAS training for employ ees, but with the acquisition of the most effec tive training medium(sJ from a host of available alternatives | source | 3 | |
| 28 | An Evaluation of the Power of the K-Sample Bell-Doksum Statistics | Edward P. Barry, Amy Ko |
sample, sample size, kruskal-wallis test | source | 6 | |
| 29 | An FSEDIT Screen Building Program to Allow Multiple Observations per Screen | Judith A. Burns, Leon T. Hairie, Thomas R. Hoffman |
Currently the SASjFSP FSEDIT procedure allows the user to edit only one observation per screen. For some applications it is more convenient to edit and/or browse a data set when all observa tions in a BY group appear together on one screen. This can be accomplished by first writing code to create a horizontal data set that contains one observation for each BY group and next rebuilding the FSEDIT screen data set. This latter step requires the user to format the screen and inform FSEDIT of the location of each variable on the screen by moving the cursor to the proper location and pressing the 'ENTER' key. For data sets with a large number of variables. this procedure can be quite time consuming. This paper describes an alternative procedure. A program has been written that 1) builds the code to horizontalize the data set. 2) creates the new screen dataset, and 3) builds the code to restore the dataset to its original vertical form |
by group processing, do over, macro variable, concomitant medications, macro definition | source | 6 |
| 30 | An Interactive Graphics Procedure for SAS Software with Application in Computer Performance Evaluation | H.M. Van Woerkom | proc step, by group processing, _n_, variable list, bar chart | source | 6 | |
| 31 | An Interface Between the Inquire Data Base and SAS System via Application Generator | Harry M. Lewin | data step, proc step | source | 5 | |
| 32 | An M-Population Covariance Stabilization: PROC VARSTB | James E. Dunn, Mike Zodrow |
Building on the notion of data based powe~ transformations, commonly referred to as Box-Cox transformations, a SAS procedure has been written for simulla-neous normalization and covariance matrix stablization in the m-sample, multi-variate case. Test runs on two quite diverse operating systems for Fisher's iris data and for a more elaborate example indicate that the iterative procedure is economically feasible in terms of CPU time to convergence as well as being relatively insensitive to starting values | covariance, proc step, likelihood function, sample, degrees of freedom, class variable, maximum likelihood estimation, normalization, regression analysis | source | 6 |
| 33 | An Overview of the SAS/REPLAY-CICS Product | Roger Williams | sample, dataset | source | 8 | |
| 34 | Analysis of a Balanced Latin Square Design Using the SAS System | Kent R. Skalland, M. Lorie Skalland |
anova, proc means | source | 3 | |
| 35 | Analysis of Covariance-Models, Strategies and Interpretations | George A. Milliken | degrees of freedom, significance level, regression analysis, ancova, covariance, proc calis, sample, proc glm, proc genmod, proc step, null hypothesis, predictor variable | source | 10 | |
| 36 | Applications of Computer Cartography for Marketing Research and Corporate Planning | Vince Bartle, John G. Blodgett, Marius Janson, L. Douglas Smith |
regression analysis, data science, sample, response variable, macro definition, proc means, proc step | source | ||
| 37 | Applying a Markov Process to Organizational Dynamics | Mark S. Edmondson | source | 6 | ||
| 38 | Applying Pattern Recognition to Validating Time Series Data for Electric Utility Load Research | Andrew E. Allen, Tom L. Johnston, Edward L. Tabakin |
time series analysis, proc step | source | 6 | |
| 39 | AURORA - Automated Rostering Application | Dick te Winkel, Erik W. Tilanus |
proc step, in= | source | 4 | |
| 40 | AUTOMAP, A Conversational Front-End to the SAS/GRAPH Product for Producing Maps | Maureen Baran, Shyh-Te Peng |
The U. S. and world mapping capabilities of SAS/GRAPHTM are used by analysts at Eastman Kodak CDmpany to display sales territory data and the results of consumer surveys. Howev er, these users of teL do not have sufficien~ SAS background to write the SAS code required to produce these maps. Thi:,; paper discusses our approaches to enable novice SAS usc~s to: Create a response data set, if necessary * using SAS/FSP Subset a response data set Generate maps Outpnt thp. maps on a shared IBM3287 printer in a VMjCMS environment. 2. Irctroduction The mapping applications at Eastman Kodak Company are usually one of the [ollowillg four types: U. S. ~LaLe and county mapping World and regional mapping Choropleth mapping with symbols User-defined territory mapping The AUTOMAP system fully supports the first two types of mapping because they are the most popular. A brief overview of ~ach of the above mapping applications follows. U.S. state and county mapping: Figure 2.la sho~s the states map |
dataset, proc step, response variable | source | 6 |
| 41 | BESTREG: A Procedure for Computing "BEST "Regression Subsets | Harold W. Gugel | Stepwi~e methods are well known to be suboptimal at determining "best" subset regressions. Com puting all possible regressions to identify "best" subsets becomes a formidable task for more than fifteen variables. However, recent advances in subset selection methods now make their appli cation for isolating "best" regressions feasible for studies containing as many as 50 or more variables. This paper describes a new SAS proce dure, named BESTREG, which implements the most recent version of the Furnival-Wilson "leaps and bounds" algorithm for "best" subset selection in regression analysis. This procedure, intended primarily as an exploratory tool for mathematical model development, provides summary statistics for the "best" and alternative "next best" regressions of all sizes. In'addition, it pro vides graphic facilities to assist the user in identifying the overall "best" model and to assess the strength of competir.g models | regression analysis, predictor variable, dataset, response variable, correlation | source | 6 |
| 42 | Changes and Enhancements to SAS/ETS Procedures | David M. DeLong, Leigh A. Ihnen, Mark Little |
proc step, missing value, moving average, time series analysis, likelihood function, sample, confidence interval, maximum likelihood estimation, proc model, autoregressive model, by group processing, estimate, proc arima, regression analysis, monte carlo, proc reg | source | 6 | |
| 43 | Choropleth Mapping with United States Geologic Survey Land Use/Land Cover Digital Files: A LIRAS to SAS/GRAPH Interface | Randy S. Bullard, David J. Cowen, Pete H. Oppenheimer, W. Lynn Shirley |
Since 1973 tbe United States Geologic Survey has been act! ve ly invol ved in mapping and inven torying the land use and land cover of the United States at a scale of 1:250,000. An important component of this program has been the creation of a digital representation of land use/land cover, hydrologic units, census county divisions and Federal land ownership maps. The USGS also developed a series of special ized computer programs, known collectively as GIRAS (Geographic Information Retreival and Analysis System), that performs all facets of digitizing, editing, inventorying and displaying the various datasets. Al though GIRAS is a powerful program set, it lacks integration into formats widely used by mapping programs such as | proc step, dataset | source | 5 |
| 44 | Clinical Trial Management Using the SAS System | Jeffrey S. Kouba, Gary D. Novack |
clinical trial, proc step | source | 2 | |
| 45 | Computing for Unbalanced Repeated Measures Experiments | Kenneth N. Berk | Repeated measures experiments involve two or more intended measurements per subject. If the within-subjects design is the same for each sub ject, and if no data are missing, then the analy sis is not very hard and there are readily avail able programs that do the analysis automatically. Huwever, if the data are not balanced, with the same arrangement for eaeh subject, then the analysis becomes much more difficult. Beginning with procedures which are not opti mal but are fairly simple, we move on to unbalanced linear model analysis, and then normal maximum likelihood procedures, We discuss ML and REML estimators for the mixed mouel and also max imum likelihood estimators for a model which allows arbitrary within-subjects covariances. We give a program which uses SAS MATRIX to do gener alized least squares based on the output of BMDPAN, which gives a maximum likelihood estimate for the covariance matrix | repeated measures, covariance, likelihood function, fixed effects, maximum likelihood estimation, proc mixed, anova, generalized least squares, missing value, degrees of freedom, proc step, proc genmod, restricted maximum likelihood, binary variable, main effect, multivariate analysis, sample | source | 7 |
| 46 | Construction of a Procedure to Convert Map Coordinates Between Geodetic and Universal Transverse Mercator Coordinates | Donald W. Sparling, Jeffry S. Yaplee |
proc step, dataset | source | 6 | |
| 47 | CPDMS-A Clinical Pharmacology Data Management System | Rita Brilla, Lorna Lyall, Lois Preis |
CpI)MS, an interactive, menu driven computer system, was developed to eliminate the manual transcriptior of laboratory test resull!> Clnd facilitate the management of this data. Tools provided by National Health Laboratories (a Revlon, Incorporated sub s id iary) , Cl is t s, and ~AS have. been integrated to create this system. USl.ng a sequential file stored on the corporate IBM 3042, a SAS dataset is created and used to update the master SAS database. This master database provides up to the minute test results needed to prepare final case report form documents | crf, clinical research | source | 7 |
| 48 | Custom Graph Axes with PROC GPLOT | Dale B. Peart | Graphs for some technical publications are required to be in a certain format. In cases where HAXIS and VAXIS options are specified, complete control of ticks, axes, and labels for graphs can be achieved by using the NOAXES option of the PLOT statement, and MOVE and DRAW options of one or more NOTE statements. The SAS-code used to create 1 eft and ri ght vert; ca 1 axes, top and bottom horizontal axes, ticks inside the axes, and custom tick labels is explained. With this information, the user then can select from all available TITLE and FOOTNOTE options to customize the axes on SAS/GRAPH! plots. Occasionally. after specifying 'HAXIS='. SAS/GRAPH will displace the r'ightmost tick label to the 1 eft as if there was not enough room to center it below the tick. The HPOS option of the GDPTIONS statement and the HSPACE option of the PLOT statement can be used to solve this problem |
proc step, standard deviation | source | 3 |
| 49 | DASD Management, An Overview of the Hows and Whys | Stanley E. Rubinson | data wrangling, elapsed time | source | 6 | |
| 50 | Data Base Design for Decision Support Systems | Gregory N. Smith | The Statistical Analysis System (SAS ~) has powerful capabilities fo~ allowing users to interact with and manipulate data to support decision ma~:ing acti vi ties. Such interactions depend upon the existence of a set of data files (a database) from which data can be selected, combir.ed and proces5ed to produce information (data useful in a speci::ic conte~:t) .'1'l:e structure of that database limits its utility for manipulating data and presenting informa~ion. Such limitations can increase programming requirements and restrict a decision makers understanding of the data. This limits the quality of decisions lildLle. This paper describes the functions, design goals and structure of a Decision Suppo~t System (DSS) database. Then, wi th this foundatiun laid, specific descriptions and cx~mples are given for use of SAS to create a database for Decision Suppo~-t Systems | proc step, cpu time, proc sort, sas function, data science, proc contents, dataset | source | 6 |
| 51 | Design Concepts for SAS Applications | Ross Merlin | proc step, proc contents, dataset, label statement, temporary dataset | source | 5 | |
| 52 | Designing Effective Graphs: Human Information Processing and Its Consequences for the SAS/GRAPH Product | Terri S. Ackerman, William A. Simcox |
Oesigning an effective graph involves more than just writing clever and sophisti cated code. Which type of graph is easiest to read? And which is most accurately per ceived? The study of human information pro cessing reveals two important limitations - sensitivity constraints and linearity con straints. These constraints are not con stant but, instead, vary with the visual dimensions used to represent the information in graphs. This fact can be u~ed to con struct alternative graphics WhlCh employ the most accurately perceived visual dimension~. We will present two such graphs that have begun to surface in the literature--~l charts and dot charts with grouping--and show how they can be constructed using | bar chart, proc step | source | 7 |
| 53 | Developing A Computer Cost Distribution Method Using the MICS Data Base | Anne N. Bluth, Jean L. Waits |
A method of allocating scarce computer resources is of prime concern to data processing managers attempting to control the s~stem and distribute the cost of these resources back to the user. Therefore a computer cost distrib ution algorithm was developed using the system resource measurements available in the HIeS data base. This algorithm was designed to discretely account for batch and time sharing resource utilization in an IBM 370/30XX environment. A background of why this is necessary and the restrictions that are_ placed on this type of algorithm are provided. It is demonstrated how SAS can determine the critical variables in an algorithm and how the corresponding coefficients are determined. Statistical methods such as the multiple stepwise regression analysis and the F-test are presented as empirical justification of the chosen approach | regression analysis, sample, correlation, cpu time, stepwise selection, proc step, null hypothesis, standard deviation, pearson correlation, proc freq, poisson distribution | source | 6 |
| 54 | Developing a SAS Training Course | Jean Ussery | If you are assigned the task of developing a SAS& training course, where do you begin? How can you decide what topics to cover and how to organize them into a logical sequence? This paper discusses a structured .technique used at SAS Institute to analyze the skills students need to learn and the development of a course based on that analysis. Materials developed may be in a variety of forms - overhead trans parencies, 35mm slides, videotapes, or text for a workbook or computer-based instruction. Since many people are concerned with develop ing an introductory SAS course, this paper uses topics in such a course as examples. The techniques, however, can be applied to the de velopment of any course, at any level of expe rience. Why Have ~ Method? Why is a method needed for course develop ment? Designing a course is like writing a com puter program. Programmers that follow struc tured design techniques usually complete assignments faster, since repetitive components are written only once | source | 3 | |
| 55 | Development of a Multi-Disciplined Environmental Data Base Management System | James R. Johnson, Marji J. Randall |
A da ta base management system has been developed to manage analyze and report the environmental data cOll'ected as' part of the Rio Blanco Oil Shale Company (RBOSC) operations on Federal Prototype 011 Shale Lease Tract C-a. Multiple master SAS data sets, comprised of data collected in five major environmental programs, are updated, edited, analyzed,. and reported using the Statistical Analysls System (SAS) progranming language. SAS is utilized because of the flexibility needed to manage large vol urnes of data and the unique graphics and analysis procedures, along with the ability to generate customized data summaries and reports | dataset, transaction data, quality control, master data management, variable | source | 6 |
| 56 | Efficient Data Retrieval-Direct Access Using the Point Option | Neil H. Howard, Linda Williams Pickle |
dataset, set statement | source | 5 | |
| 57 | Estimating a Transformation Between Projections Using Least Squares | Paul W. Fingerman, Marcia A. Tolbert |
correlation, sample | source | 3 | |
| 58 | Estimating Power for Tests of the Multivariate General Linear Hypothesis | Keith E. Muller, Bercedis L. Peterson |
Although methods for exact power calcula tions for tests of the multivariate general linear hypothesis do not exist~ convenient accurate approximations are available for three commonly used multivariate test statis tics: Wilks' Lambda, Hotelling-Lawley Trace, and Pillai's Trace (Muller and Peterson, 1983). In thi s paper two computer programs are presented which calculate these approxi mate powers. One program performs "prospec tive" power analyses in which powers are calculated for tests on hypothesized models. The other program performs "retrospective ll power calculations on observed test statistics | sample, sample size, power analysis, dataset, proc step, degrees of freedom, covariance | source | 5 |
| 59 | Expert Based Training of SAS Users: Getting Productivity Fast | Nelson R. Pardee | data wrangling, sas programmer | source | 3 | |
| 60 | Financial Statements and Ratios from CompustatTM Accessed and Displayed Using SAS Software | J. R. Byington, M. T. Dodd, Mary S. Hirsch |
The compustat R data base distributed by Standard and Poor's Compustat Services, Inc., was input into a SAS data set with an updated version of macro SPYR (SAS/ETS). Appropl'iate inuustrial annual data items were selected to produce the financial statements as indicated by Compustat. Financial ratios were also calculated for the selected company, and industry means of these ratios were determined. The data were transposed, all data items were labeled and footnotes were included so that pertinent information was dis closed in a meaningful format. This is a practical method for businesses and classrooms since the user supplies only the company number-or ticker symbol to obtain the financial statements and ratius. The::;e SAS pru grams thus allow straightforward access to and conventional display of compustat annual data |
dataset, input statement | source | 6 |
| 61 | Forecasting Customer Air Conditioner Loads and Load Drops for Electric Utility Load Management Program Planning | Marjorie R. McRae, Richard M. Scheer, Bruce A. Smith, Louise Weiler |
correlation, autocorrelation, proc reg | source | 4 | |
| 62 | Forecasting Industrial Production in 1981-1984 | Carlos T. Bailey | interest rate, confidence interval | source | 8 | |
| 63 | Format 1-2-3, A SAS Software to LOTUS 1-2-3 Interface | Valerie Van Langen, Dennis Van Langen |
proc step, proc print, proc univariate | source | 6 | |
| 64 | FSGRAPH-A Combination of the SASIFSP and SAS/GRAPH Products for Developing Graphics | Roger D. Muller | A system has been developed to generate SAS code for all procedures in the SASjGRAPII product from data entered in SAS data sets on screens in PROC FSEDIT of the SAS/FSP package. Data entered on screens includes such items as data set names, variable names, axis colors, labels, etc. The system permits for rapid creation of graphics during interactive SAS sessions, with provisions made for editing and storing both the input SAS data sets and the resultant generated SAS code. Provisions are also made for storing and replaying the graphic images using GOUT data sets and PROC GREPLAY. The system is particularly useful for decision support applica· tions where users with minimal programming knowledge desire to use SAS/GRAPH as an exploratory data analysis tool on previously created SAS data sets. An extenuation of the system permits for use of selected pro cedures on nongraphics devices such as nongraphics terminals and printers |
proc step, dataset, horizontal bar chart, bar chart, sample, macro definition, sas programming | source | 8 |
| 65 | Full Screen Interface for Line Graphs | G. Nicholas Lauer | dataset, line chart | source | 3 | |
| 66 | Generating Bivariate Exponential Distributions via SAS Macros with Application to Repeated Measures Designs | Melvin Moeschberger, P. K. Tandon |
nonparametric test, type i error, exponential distribution, significance level, sample, sample size, macro definition, correlation, main effect | source | 4 | |
| 67 | GENISYS-A Generalized Inventory System for Clinical Trial Data | Pamela A. Frueh | Patlent accountability is of prime importance in the analysis of clinical trial data for a New Drug Applica tion (NDA). The Generalized Inventory System (GENISYS), in use at Hoochst-Roussel, is a | data science, proc step | source | 5 |
| 68 | Getting Started with the SAS/I MS-DUI Product | Cheryl Turney | source | 9 | ||
| 69 | Graphic Representation of Logistic Regression Models | John K. Gohagan, Donald G. Sessions, Edward L. Spitznagel |
roc curve, regression analysis, logistic regression, logistic distribution | source | 4 | |
| 70 | Graphical Analysis for Telecommunications Engineers | Douglas Sherman | This paper describes the COMSYM software system designed for IBM's Internal Voice Network by the Network Management Systems Development depart ment. This system is used by telecommunications engineers to design and analyze IBM's internal voice network. The system incorporates different languages to provide a comprehensive reporting tool for tele communication engineers. Thl'! paper will focus on the graphics subsystem which is written using SAS and SAS/GRAPH. This system is designed to be used by non-technical people. Front end pan els (screens) were designed to prompt the user for parameters which determine the input records used as well as the plot produced. The large amounts of data associated with call detail records, performance records, and utility records are easily handled by SAS and can be displayed effectively using SAS graphics. The ability to generate graphs and isolate certain information is required in order to manage a complex voice network |
proc freq, bar chart | source | 6 |
| 71 | How the SAS System is Used by Division of Cardiovascular Surgery at Toronto General Hospital | R. J. Baird, B. S. Goldman, E. J. Kuzin, L. L. Mickleborough, H. E. Scully, Cathy P. Tong, R.D. Weisel |
SAS system was chosen as the software pack age for implementing data bases at the Division of Cardiovascular Surgery at Toronto General Hospital. Among the many retrospective and pro spective studies done by the Division, four on going data bases were established between the period of 1980 to 1983: Metro Toronto Open Heart Data Base To collect clinical, operative and post operative data on each patient in order to document indications and results in the city wide practice of Cardiovascular sur~ery. Physiological Data Base To examine ~ocardia1 performance as a result of various techniques used on patients during and after cardiopulmonary bypass. Intra-Aortic Balloon Pump Data Base To document each patient that requires bal loon counterpulsation. providing information on patient survival and complications. Dual Chamber Pacemaker Data Base To monitor the on on-going follow-up and clinical importance of the new technique of dual chamber pacing. Standard frequency analyses | standard deviation, regression analysis | source | 2 |
| 72 | Human Factors Engineering in SAS/FSP Applications | Kirk Paul Lafler | The primary focus of human factors engineer ing is to design for human use, that is, to be functionally effective. With a greater number of SAS full-screen applications being developed, many designers have neglected to incorporate human factors engineering into their designs. One reaRon for thiR neglect could be that be cause human factors engineering is e_ highly in terdisciplinary field, the relevant literature is scattered. Or there may be a deeper reason, typically, a lack of understanding exists in the design of man-made tools for human use. These and numerous other suggestion~ are contributed to assist in the design of human factors engin~ eering in SAS/FSP applications |
error message, likelihood function | source | 2 |
| 73 | Improving Programmer Productivity with Coding Conventions | Anne M. Asher, Migdalen C. Eley |
proc step, do loop, if-then-else, conditional do loop, do over, subsetting if | source | 6 | |
| 74 | Initial Experiences with the SAS System for Minicomputers | C. David Hardison, Edwin D. Suckstorff |
This paper describes the experiences with the SAS System on a DEC VAX-Il/7eO at Lilly Research Laboratories. Same differences between the SAS System under VAX/VMS and the SAS System under VM/CMS or MVS are described, and examples are given which illustrate the use of the SAS System: 1. to analyze VAX system performance data, 2. to analyze and generate reports from data collected on remote DECnet nodes, 3. to move SAS applications between the VAX/VMS environment and the IBM VM/CMS environment with PRoe XCOPY. THE SAS SYSTEM FOR MINICOMPt~S FACY~GE The Program Products available for minicomputers differ from those available for IBM computer systems. The major differences between the SAS System for minicomputers and the "standard" SAS System are summarized in Figures 1 and 2. At the time of this writing, the SAS System for minicomputers only includes SAS/BASE and SAS/GRAPH, plus Display Manager, which is a new development tool available only with the SAS System for minicomputers |
dataset, data science | source | 6 |
| 75 | Interactive Psychological Testing Using the SAS System: A SAS Program to Administer and Score the MMPI | Joseph M. Antonello, Robert M. Hamer |
dataset, raw data | source | 5 | |
| 76 | Interactive SAS Programming | Harriet E. Brichta | this paper is interactive SAS Specifically, a technique is developing programs which enables users, unfamiliar with SAS syntax, to create and execute SAS programs. The steps involved in interactive SAS programming are summarized, and €l:XamplAs arA presented which employ an IBM software product, EXEC2, as the pre-processor. This presentation should provide a useful comparison for SAS program generator programs (discussed in Chapter 12 of the SAS Applications Guide), and an alternative strategy in some cases | time series analysis, dataset, sas programming, time series data | source | 6 |
| 77 | Interactive SAS Use in Batch Mode: Making the SAS System User Friendly in a ROSCOE Batch Environment | S. David Riba | While there has been much discussion about the use of SAS in an interactive environment such as TSO. users of SAS in batch mode have a variety of problems that need to be addressed. At E-Systems we have developed several methods that have proven effective in minimizing these limitations. This paper will re view our experiences with modular pro gram design. ROSCOE RPF. and in-house SAS support | dataset, data science | source | 3 |
| 78 | Interfacing SAS Software with the SCA System: the SCALINK Procedure | Lon-Mu Liu, Houston H. Stokes |
The SCALINK procedure provides a link whereby SAS jobS can uti lize the extens'lve time ser'les capability of the SCA statistical system | proc step, correlation, sample, time series analysis, dataset | source | 6 |
| 79 | Interfacing SAS/GRAPH Software with Graphics Hardware | Mike Kalt | batch job, proc gplot | source | 8 | |
| 80 | ISPF Interplay with the SAS System | Vladimir Svirsky | source | 5 | ||
| 81 | Keyed-Access to SAS Data Sets | Alec Ramsay | This paper discusses the storage space - comput ing time tradeoff inherent in many SAS data sys tems and presents a method for achieving keyed access to SAS dllta sets that eliminates the tradeoff | dataset, macro variable | source | 4 |
| 82 | Linear Goal Programming Using the SAS/OR Product | Marc-David Cohen | Many problems that arise in the practice of management science require decision making in the feee of multiple objective::;. Usually these | linear programming, phase i, phase ii | source | 11 |
| 83 | Logistic Regression and Associated Diagnostics using PROC MATRIX | Thomas F. Devlin | logistic distribution, regression analysis | source | 6 | |
| 84 | Making Systems Portable | Richard A. Usanis | source | 8 | ||
| 85 | Managing a Data=Base of Permanent SAS Data Sets | Joshua Sharlin | dataset, proc step, proc sort, audit trail, proc copy, put statement, update statement, proc append, sysdate9 | source | 9 | |
| 86 | Managing Human Resources Information: Using the SAS and SAS/FSP Products in the CMS Environment | Joel F. Achtenberg, Douglas G. Fishel, Paul M. Roesti |
A Human Resources Management System (HRHS), under development for the Personnel Department at the Washington University School of Medicine in St. Louis is described. The primary function of the system is to match applicants with employer requirements. A pool of applicants is created, composed of potential employees. The pool contains applicants' skill levels for 211 skills, in addition to standard application data. Employers specify minimum levels acceptable for these same skills on job requisitions. The applicant pool is searched using a requisition to identify candidates with skills meeting or exceeding the requisition levels. Personnel staff may interactively relax or strengthen the search criteria. Potential employees are identified, and records of all referrals to employers and employers' decisions are kept for each applicant. When applicants are hired, they are added to an Employee Skills Data Bank | proc step, variable | source | 4 |
| 87 | Mechanization of the Income Tax Accrual using SAS Software | James H. Sorenson | Once each month, accountants in public utilities across the nation put pencil to paper and attempt to prepare workpapers to support the monthly income tax accrual. This task can become quite frustrating as earnings estimates, schedule "M" projections, ITC forecasts and numerous other variables change constantly throughout the year. Accountants have tradi tionally tried to solve the problem of docu menting a complex series of tax calculations by finding the best workpaper format. Certainly a well-organized, logical approach to the problem will do much to make the job easier. Even the best workpaper format, however, will do little to recompute the calculations wheo an assump tion is modified or review the workpapers for math errors. At San Diego Gas & Electric (SDG&E), the tax accrual was becoming more and more com plex | dataset, data step | source | 6 |
| 88 | Microcomputers and SAS Software in a Distributed Human Resource Information System | Ervin H. Young | named input, data step | source | 3 | |
| 89 | Modeling Congenital Cardiac Surgical Caseload Using the SAS/ETS Product.. | Steven P. Davis, Ronald C. Elkins, Mary M. Lane, Cynthia K. Murray |
confidence interval, proc freq | source | 6 | |
| 90 | New SAS Survival Procedures | David M. DeLong | response variable, out=, rank test, dataset, proc lifereg, likelihood function, proc step, regression analysis, confidence interval | source | 3 | |
| 91 | New SAS Tools for Data Base Management | N. D. Prabhakar | raw data, dataset, macro definition, keyword parameter, %macro parameters, by group processing, missing value | source | 3 | |
| 92 | New SAS/GRAPH Features | Jack Bulkley | sample, proc gchart | source | 6 | |
| 93 | NODIS Procedures for Demographic Analysis | Mark J. Salling, Janice S. Tobin |
variable list, missing value, proc step, proc print, dataset, by group processing | source | 6 | |
| 94 | Nutrient Data Base Systems as Educational Tools for Allied Health Sciences Professionals using the SAS/FSP Product | John W. Corliss, Wayne W. Crutcher, David P. Menapace, Mark A. Simmons |
Computerized nutrient data base systems are available and can be utilized as instructional tools by dental. medical and other students in the allied health sciences in the study of nutritional science. \Hth these data banks, students can develop skills in the compilation and the analysis of patient food consumption histories, in the calculation of nutrient composition of foods. in the design and evaluation of individualized menllS for meeting specific nutrient standards and objectives in preventive health care, and in the utilization of nutritional assessments for the development of clinical treatment strategies. In the summer of 1982, such a. data base system was designed and implemented by the academic computing services department (ACS) to support research, education and clinical applications in nutritional science at the Loyola University Medical Center (LUMe) | batch processing, dataset | source | 6 |
| 95 | On Generating Statistics, Formatting Results and Efficiency | Ingrid A. Amara, Gary G. Koch |
proc step, proc print, output, proc univariate, proc printto, randomization | source | 6 | |
| 96 | On the Limitations of Akaike's Information Criterion and Its Use in PROC STATESPACE | Jim Chavern | This paper is aboo t FROC STATESPACE and how this program uses Akaike's Information Criterion. Basic criticisms of PROC STATESPACE are: 1) If the user doesn-t pick the form himself, FRoe STA1ESPACE picks the model form of a time series using Akaike's Information Criterion. 2) Akaike's work well. Criterion It tends often doesn't to overes t ima t e the number of parBmeters in a time series model. 3) The penalty for overestimating the number of parameters in a time series equation can be Ilubstantial, in terms of the mean square error of prediction. 4) There is no such thing as an "auto matic statistical method" any more than there is an automatic method of think ing |
time series analysis, likelihood function, proc step, sample, autoregressive model, mse | source | 6 |
| 97 | On the Study of Clusters with Factor Analysis | Eric Deaton, Richard Israel |
factor analysis, correlation, proc step, sample | source | 6 | |
| 98 | Panel Discussion | John C. Boling, Mason B. Nichols |
sas consultant, macro facility, merge statement, data wrangling, sample, proc step, sas base | source | 17 | |
| 99 | Panel Discussion of SAS Users under the AOS/VS Operating System | Jon Fleig, Jerry Harber, Steve Sashihara |
data wrangling, decimal | source | 4 | |
| 100 | Panel Discussion of SAS Users Under the VMS Operating System | Louise Horny, Kevin Mallory, Gerald Perry, Brad Reese |
proc step, cpu time, sample, dataset, regression analysis, proc genmod | source | 7 | |
| 101 | Power Analysis for Linear Models: The Time Has Come | Virginia I. Lohr, Ralph G. O'Brien |
The purpose of this session is to show you how to use the SA~ System 3 to methodically ex amine the statistical power of your linear-mod els analysis. The purpose of this particular paper is to set the stage for the other papers, which contain the core of the technica~ materi al. The next paper, Lohr and O'Brien (1984), handles univariate linear models. Fol~owing that, Muller and Peterson (1984) extend matters to the multivariate case. We are pleased and grateful that Rudolf Freund has agreed to discUSR our works. Of related intfl:rest is a poster pa per, Peterson and Huller (1984), which presents specific SAS code for the multivariate computa tions. To set the stage for "Practical Methods for Power in Linear Models," we will now address these areas: Noncentrality parameters, noncentral dis tributions, and power. An example of a power analysis. Power analysis in consulting: a task for ~arly researcher-statistician communica tion. Power analysis and teaching | power analysis, sample, anova, degrees of freedom, proc glm, proc step, null hypothesis, f-test, sample size, student t distribution | source | 7 |
| 102 | Power Analysis for Multivariate Linear Models: New Results and SAS Software Make It Practical | Keith E. Muller, Bercedis L. Peterson |
General approximatlons have been known for a decade for computing power in multivariate linear models tests with Wilks' lambda~ Hotelling-lawley Trace and the Pillai-Bartlett Trace (Lee. 1971. Sugiura and Fujikoshi. 1969). All use complicated mixtures of noncentral chi square variates. Recent results (Muller and Peterson. 1983) based on single noncentral F variates provide far more convenient approxi mations. The basic ideas needed to implement the new approximations are reviewed. A direct appro~ch, using matrix manipulations, and an indirect approach, using linear models programs, are sketched. I NTRODUCTl ON Moti vat; on Power calculations in testing the mUlti variate general linear hypothesis have pre sented difficult problems. Although not well known (despite publication in some of the most widely read statistical journals), practical asymptotic approximations have been available for a decade. More convenient methods have recently been proposed | correlation, proc cancorr, multivariate analysis, degrees of freedom, covariance, response variable, repeated measures, sample, power analysis | source | 6 |
| 103 | Power Analysis for Univariate Linear Models: The SAS System Makes It Easy.. | Virginia I. Lohr, Ralph G. O'Brien |
sample, power analysis, degrees of freedom, proc step, main effect, sample size, proc glm, proc genmod | source | 6 | |
| 104 | Prechecking SAS Syntax with a Microcomputer | Jeffrey S. Bass | proc step, dataset | source | 2 | |
| 105 | Predicting Prison Population With the SAS/ETS Product | Steven P. Davis | time series analysis, correlation | source | 6 | |
| 106 | Preliminary Transformations in Time Series Modeling | John C. Brocklebank, David A. Dickey |
unit root, time series analysis | source | 6 | |
| 107 | Problem Tracking at SAS Institute | Margaret L. Adair | sas consultant, proc step | source | 7 | |
| 108 | PROC ARKMAP: A Low Cost Method of Presenting Geographical Data in an Educational Environment | Dennis W. King, Richard M. Smith |
by group processing, numeric variable, character variable, standard deviation | source | 3 | |
| 109 | PROC CPM | Radhika V. Kulkarni | proc step | source | 10 | |
| 110 | PROC FSCALC in the Estimating Environment | B. J. Collins, Robert T. Grant |
proc step, missing value | source | 3 | |
| 111 | PROC GLM-Rearrangement of Output | Karen L. Bourn | proc step, least squares mean, proc glm, proc genmod, clinical trial, anova, proc printto, repeated measures | source | 6 | |
| 112 | PROC JACKREG-A SAS Procedure for Jackknife Regression | Daniel M. Chilko, Panayiotis Hambi, Gerald R. Hobbs |
PROC JACKREG allows the user to calculate robust estimates of regression coefficients in simple or multiple regression problems by use of the Jackknife method. See Mosteller & Tukey for a good description of the Jackknife. PROC JACKREG may also be used as a validation tool. The Least Squares Regression equation is validated by comparing each observation to the equation generated when that observation is left out. Output includes the individual regression coefficients at each step of the leave-one-out process with their corresponding R-square values. A further table of the final Jackknife coefficients and the ordinary least squares coefficients. with their estimated standard deviations, t-stati8tic8 and P-values are a180 displayed. Output data sets may be created optionally upon request. One of these data sets may contain the individual Regression coefficients at each step of the leave-one-out proeess (OUTEST=) |
regression analysis, proc step, predictor variable, proc reg, output, proc calis, response variable, dataset, by group processing, sample, data cleaning | source | 6 |
| 113 | PROC QRPT: Procedure for the Estimation of ED50, Relative Potency and Their Fiducial Limits in Quantal Response Assays | Nguyen V. Dat, Ronald G. Fleming, Sue Huang |
PROC QRPT procedure is a SAS procedure which provides estimations for ED50 and Relative Potency and their fiducial limits in a Quantal Response Assay_ This SAS procedure is written in FORTRAN and is geared to the convenience of the user in the SAS enviroment. 1. The frequency counts must be sorted by preparation and dose (BY VARIABLES also require sorting. if they are being used). This program estimates a 50~ effective dose (ED50) and relative potency with their 95% fi duci a I lim; ts wi thi n the framework qf Fieller's Theorem from a joint probit X2 statistics for testing the similarity of the probit regression lines required for attaining parallelism. PROC QRPT is very user-oriented and can be tun easily by usi ng SAS data sets. INTR DOUeT! ON We would like to thank Dr. Richard Lam for contributing the original FORTRAN program in Which this PROC is founded. If you would like a copy of thi s PROC, please ~ontact Ronald Fleming at Squibb Corporation. 'Princeton. NJ |
proc step, confidence interval, proc freq | source | 4 |
| 114 | PROC SPFLINK: A Direct Interface Between the SAS System and ISPF | Luigi Muro, Larry Shetzer |
The relational nature of a SAS database often puts it at an advantage over a hierarchi cal database for ease of access and maintenance. However, continually updating a SAS database can slow turnarounrl, particularly as the size of the database increases. To reduce access time we developed a procedure which transfers data directly from a SAS database to an ISPF table. This paper assumes the availability of ISPF (Interactive System Productivity Facility) and the 'Dialog Management Services' required to develop an interactive application under ISPF | dataset, proc step, numeric variable | source | 6 |
| 115 | PROC SUMMARY as the Basis for Efficient Analyses of Large Files | Carol Leininger, Juliana M. Ma |
proc step, binary variable, proc tabulate, proc means, nominal variable, class variable | source | 4 | |
| 116 | Procedure Writing: How to Implement Custom Procedure Statements | Richard L. Gimarc | This paper presents a technique for im plementing custom procedure statements in user-written SAS procedures. The steps required to implement a custom procedure statement are presented with sufficient detail such that potential procedure writers will be able to design and imple ment user-friendly procedures with non trivial supplementary procedure state ments. The design and implementation of a custom procedure statement for PROe M204 is used as an example of the tech nique presented | if-then-else, proc step, sample | source | 7 |
| 117 | Progress and Prospects | Richard A. Usanis | proc step, error message, dataset, macro facility, missing value, macro definition | source | 6 | |
| 118 | Proposed U.S. Passenger Car and Light Truck Sales Forecast Model | John Janakiraman, Stuart J. Olson Jr. |
predictor variable, response variable, correlation, regression analysis | source | 5 | |
| 119 | PUMPCSR: An Interactive Data Management System for Customer Service Requests of Infusion Pumps | Peter Chen | error message, sample | source | 4 | |
| 120 | Quality Control Through Random Error Generation | William M. Taylor | Manually keyed data may need to be visually checked to achieve a high level of precision. The visual checking process involves comparison of the original document against a listing of the data file. Various problems are encountered in monitoring the visual checking process to ensure adherence to quality standards. One approach utilizes the injection of "errors" in listings to be checked. Errors can be randomly assigned to variables and observations. Error frequency can be modified. A sample program with supervisor key to error location shoW's utilization of this technique. Unbiased measures of data checking procedures can be obtained as well as measures of data accuracy | proc freq, sample, dataset | source | 4 |
| 121 | RANKPLOT: A SAS Macro for Generating Line Printer Scatterplots of Points and Point Labels Without Overprinting the Points and Labels | Warren F. Kuhfeld | proc step, macro definition | source | 6 | |
| 122 | Rapid Response to the FDA Questions on an NDA | Kristin K. O'Connor | When questions are received from the FDA about a pending New Drug Application, they must be answered as quickly and efficiently as possible. This task is made easier with the use of SAS. This paper deser ibes the use of PROe GLM and PROC TABULATE to answer two questions recently raised by the FDA about a pending Ne\l Drug Appl ieatian |
proc genmod, proc tabulate | source | 4 |
| 123 | Repeated Measures Analysis in PROC GLM | Phil Spector | repeated measures, anova, proc glm, proc model, repeated measures anova, response variable, correlation, residual, covariance, proc step | source | 10 | |
| 124 | REPREG: A Repeated Measures Regression Program | Randy L. Carter, James T. Love |
regression analysis, macro variable, generalized least squares, sample, proc step, null hypothesis, covariance, confidence interval, predictor variable, sample size | source | 6 | |
| 125 | Residential Energy Forecasting: A Pragmatic Application of Box-Jenkins | Michael F. Jacob | This paper applies Box-Jenkins time series analysis to the problem of modeling and forecasting monthly residential sales of electricity. Using the SAS/ETS ARIMA procedure, the identification, estimation and diagnostic checking stages are out lined and forecasts presented for an ARIMA transfer fUnction model. Due to the strong dependence of residential electricity usage on weather conditions, transfer functions are specified for the two input variables, heating degree days and cooling degree days. In order to account for the time varying nature of the weather parameters, the input series are precon ditioned by the saturation of weather sensitive appliances. A further input, the number oj resi dential customers, is incorporated through a use per-customer specification of the dependent variable. Once the final model has been selected, it is used to prepare a twelve month forecast of residential sales during the year 1983 |
time series analysis, seasonality, sample, residual, stationarity, correlation, data cleaning, seasonal difference | source | 6 |
| 126 | SAS Applications to a Spring Wheat and Sunflower Yield Modeling Project | Paul G. Held, James Knuteson, D. Patterson, Michael Ulmer |
do over, data science, proc reg, regression analysis | source | 4 | |
| 127 | SAS Data Base Systems for Research and Development, Analysis and Data Collection | Lee M. Schwartz | Database systems~tilizing mini or personal computers as SAS1U program generators are powerful and cost-effectIve data collection and analysis tools. These database systeas aeet the needs of a Researcn and Development environaent. The capabilities of the database systems include data collection. maintenance. analysis. and reporting. All systsas can be supported by a general analysis package ghich features subsetting and the coaaon SAS PAOCs. Database operation eliainates the need for users to knog the SAS language and editing techniques. TABLE Of CONTENTS | data wrangling, proc step | source | 6 |
| 128 | SAS Execution Panels for Users | William R. Greenwood | At General Dynamics, the goal of GD INFONET is to aid non-data processing personnel in producing their own ad-hoc r~ports and stand alon~ systems. The SAS menu allows users to execute SAS interactively or submit d batch job without needing to kno~ how to write their own CLISTS or JCL. If an error occurs~ a message is written to the menu. An on-line tutorial is available to show the user how to fill in the menu properly. To receive the tutorial, the user just presses the PF HELP KEY | batch job, raw data | source | 5 |
| 129 | SAS Functions Useful for Probability Computations | N. D. Prabhakar | binomial distribution, probability distribution, poisson distribution | source | 2 | |
| 130 | SAS Macros for Sample Size and Power Calculations | Erik J. Bergstralh | sample, sample size, alternative hypothesis, clinical trial, median survival, macro definition | source | 6 | |
| 131 | SAS Methods for the Rank Regression Model with Missing Values | R. Michael Rubison, Michael K. Thran |
Recent trends in regression analysis have led to a variety of robust estimation procedures. The topics of influential observations. residual diagnostics and outlier detection have received a great deal of attention in both the theoretical and applied literature. Selected diagnostic techniques have appeared in the statistical software; however, most robust regression methods are not readily available to the general user. The rank regression Illodel provides a robust alternative to its parametric counterpart and is easily implemented within SAS. This paper illustrates the general properties of the rank regression model and the potential problems aS50ciatp.d with its use in the presence of missing values. Under these conditions, the different ways in which the model estimates can be derived using PROC RANK, GLM, REG, CORR, and ~~TRIX are discussed and compared |
missing value, regression analysis, correlation, predictor variable, spearman correlation, proc step, response variable | source | 5 |
| 132 | SAS Processing Techniques | Tom Krysak, David Riave |
The Group for Revenue Analysis and Special Projects in the New York City Health and Hospitals Corporation has used SAS extensively in developing applications for monitoring the Billing and Accounts Receivable activity of its 16 hospitals and health care institutions. The data for these analyses and reports are taken from transaction tapes generated by a major data processing service bureau which serves hospitals throughout the country. Reading this data was complicated by multiple record types within a file, variable length records, a variety of different storage formats and reading data from print image files. In the course of this paper we will discuss techniques that were employed to ove-rcome these obstacles in the hope that others, may find them useful in their SAS applications. Topics include Create multiple input statements using trailing @ sign and conditional logiC statements. OUtput single observations from variable and quasi-variable length records | length statement, conditional do loop, input statement, character variable, numeric variable, substr function | source | 6 |
| 133 | SAS Programs for Reporting the Results of Bioequivalence Trials | James T. Crowe Jr., Thaddeus J. Malicz |
Federal regulations require that pharma~eutical companies demonstrate the bioavailability and bioequivalence of new products and of new form ulations of existing products. Bioavailability and bioequivalence are established by clinical trials in which concentrations of drug in plasma, urine, or both measured after adminis tration of the compound. The consistency of the type of data and the experimental designs makes it possible to develop a system of SAS programs to analyze and report the data. The modularity of these programs makes adaptation from one trial to the next a simple procedure | anova, clinical trial, data science, dataset | source | 3 |
| 134 | SAS Software, the Gateway to Microcomputer Applications | J. Harvey Trimble Jr. | The dF?r"I)alld t{".) couple the intel'"'iH:t:ivH pewit;"r cind easy to use appi ication ptl.c:kage':> (.)f microcampu"l:er-s wi t.h Lhe 1 aTqe data nases and cnmplxt<..'ltional f)(ll"ler~ of IB!'l mainframes has qiven t'"ise to thp annollnc:ement of several pr'oducts which claim to link ttlE? tt.oJo comput.p.I'"'s. These rH"oduct!;:; pl'"ovide vat"ioLls kirHis of corulF-.>ctions which in some t:a~>e unJ.)' suppor-ts tr'cHlsfer' of inTIJF'mation from IBt1 mainframes t(:J fili(.:rocompt.ltt'~r-s but not back to t.he maillft-amr=-. A more f.le->:ibl!:? ,,,,It..et'"native, one which is 1'10-:. tied to any specific micr'ocIJlJlputer i:\pplil.:ation packaqe Ll"-:;es SAS as Ule qCl:i:eway between tiif'? mJ.cr(Jcomput8t~ and t.he mainfr-ane. TI')is dol t:?r-ndti. Vf2 is d 5A~:; 82 1"-1;;tc.:r"'o Pt'"Qcedut-e named TRANSFER. TRfC.)NSFER is bui 1 t u~on the p()"'H~r"ful data tt·ansfClrl!l<:\·tion and info,.-fir,:!tion displclY capabilities of SAS [1. ] I .o\n(1 provl | source | 6 | |
| 135 | SAS/GRAPH Applications for Plotting Saline Soil Profiles | Paul G. Held, Alex Maianu |
A clear appreciation of saline soil profiles may be made by graphically viewing the anions and cations plotted with depth. Using the PROG GPLOT procedure, with the fill option, a cross-sectional view of the quantity and depth distribution of every anion and cation is ob tained. Through viewing the proportion of every anion and cation wiLh Lhe 50il depth I more def inite conclusions can be made on the genesis and properties of salt-affected soils |
proc step, proc gplot | source | 5 |
| 136 | Saving Time with the SAS Display Manager | Philip James Busby | sas log, proc print | source | 23 | |
| 137 | SEQURAN-An Interactive Sequential Randomization System for Clinical Trials | Estelle M. Fondy, Jeannette Y. Lee |
TreatmEnt groups with similar prognoses arE! essential in obtaining valid results in a clinical trial. Sequential treatment assign ment methods which consider known prognostic factors have been developed to assure balanc ed treatment groups. These methods are adap tive in that the treatment assignment for the nth patient depends on the treatment assign ments and prognostic characteristics of the n-l patients already enrolled ill the study. However, treatment assignments cannot be made in advance. This provides a unique oppor tunity for utilizing the interactive capabil iti~s of the eomputer. In this paper ~ W~ present SEQRAN. an interactive system for assigning treatments in a clinical trial using a sequential randomization approach. SEQRAN performs two vital functions: screening patients for eligibility and assigning treatments to eligible patients who enroll in the study. Several features of SAS are critical to SEQRAN. SAS/FSP and eMsl SAS make SEQRAN interactive and user friendly |
clinical trial, randomization, dataset | source | 6 |
| 138 | Solving the IBM Graphics Problem | Bill Manos | source | 2 | ||
| 139 | Some Preliminary Tests in Repeated Measures Design | M.F. Huque, Harji I. Patel, T. Roy |
This paper deals wit_h the analysis of variance or covariance in a repeated measures design with at most two grouping factors and one repeated measures factor. Using the MATRIX procedure, a macro is written to test the hypothesis of equal cell VC (variance covariance) matrices and the hypothesis of equal cell VC matrices with a common matrix having Type-H structure. The macro also computes e, the multiplicative adjustment factor for the degrees of freedom for the F-tests in the analysis of a mixed model_ | degrees of freedom, covariance, repeated measures, sample | source | 6 |
| 140 | Stagewise Experimental Design Techniques for Nonlinear Models | Thomas J. Bzik | proc nlin, response variable, regression analysis | source | 9 | |
| 141 | Starting a Local SAS Users Group: Some Issues and Guidelines | Jim C. Knoop | proc freq, sample | source | 7 | |
| 142 | Statement Style Macros | Donald J. Henderson, David L. Kuhn |
sample, sas log, macro definition, error message, macro variable, macro loop, data step, proc print, proc step, macro facility, %if-%then-%else | source | 10 | |
| 143 | Statistics Automation: Using SAS Macros to Produce Report-Ready Statistical Analysis Tables | Walter W. Offen | Experimental designs are often similar or identical to designs of studies previously analyzed. Fnr example, in large-scale clinical trials. analyses of individual centers or other subgroups of patients are often required. For this reason, SAS macros have been written to analyze data. These are general in the sense that they can be readily re-applied to a later study with the same basic design. The macros require only that the SAS dataset be in a specified conventional form, and that several parameters be defined prior to invoking them. Hence, they are quite simple to use. The output is a statistical table. complete with summary statistics, tests of hypotheses. and confidence intervals. These tables are suitable for direct insertion into statistical reports. The purpose of this paper is to familiarize the reader with the technique of writing these macros, and to point out the advantages the macros offer. A relatively simple macro is presented as a demonstration of the technique | anova, proc step, macro definition, proc means, confidence interval, proc format, significance level, sample size, data science, sample, clinical trial | source | 5 |
| 144 | STRATI F: A Procedure for Producing Stratified Variables | M. G. Sri Ram | proc step, class variable, numeric variable, missing value, variable list | source | 5 | |
| 145 | Technical Exposition of Graphical Tools in Survival Analysis | Gregory G. Enas, C. David Hardison, Frank W. Rockhold |
proc phreg, correlation, spearman correlation | source | 6 | |
| 146 | Techniques for Applying SAS Software to VM/CMS-Based Business Systems. | Laura S. Vendeland | The business programmer/analyst possesses a very powerful tool when the flexibility and wide appli cability of SAS software is combined with the quick, interactive mechanism of the VM/CMS operat ing system. By weaving eMS EXECS, panels, and macros around SAS programs and utilities, easy-ta-use business systems can be built to opti mize the capabilities of both tools for the programmer and the end-user. This paper disu5ses techniques for implementing SAS utilities under VM!CMS to build araphics and reporting systems, to use SAS programs as both a pre- and post-processor to a batch job, and to write a complete trans actions file from a SAS/FSP update session |
batch job, dataset, proc step | source | 6 |
| 147 | Tennessee's Federal/State Tax Comparison | Virginia N. Adams | proc step, proc copy | source | 3 | |
| 148 | The Construction of a Standardized Analysis File From Multiple Inputs | Zelda B. Cohen | sample, dataset | source | 6 | |
| 149 | The Information Center at BBL Microbiology Systems | Patricia W. Botelle | data wrangling, proc step | source | 5 | |
| 150 | The Role of the SAS System in the Information Age | David R. Dolkart | New hardware and softvvare produc1s come into the marketplace at a phenomenal speed. It is hard to predict which of these procucis will survive the f"erce competition in the months and years to come. INi!h this rapidly changing environment, SAS InstitJte Inc. ~as a formidable task. That task is to support a balanced approach to the infor"Tlation age. SAS Institute's track record ;s strong eviderce that this balance will go well beyard condensing its mamframe software for microcomputers. Footnotes 1. SAS is u regblereu truuemu·k of SAS InstitLte Inc., Cary, NC, USA. ? IP.M Personal Computer XT/370 is a reg:stered trademark of International Business Machines Corporation. 3. Guiliano, V. E., "The Mechanization of Office Work," Scienlifi::: American, Vol. 247, No.3, September 1982, p. 164. 4. Ibid., p. 160. .5 Mouton, P. cnd Colfin R. Personal Computers, Answering User Needs, American Management Association, 1983, p. 11. 6 | source | 1 | |
| 151 | The SAS System and the Microcomputer | P. L. Olympia | statistical software, regression analysis | source | 5 | |
| 152 | The SAS System as an Information Center Tool | Patrick R. Timmins | sas consultant, data wrangling, sas output, macro facility, sas programmer, merge statement, sample, sas base | source | ||
| 153 | The SAS System Can Replace Traditional Programming Languages | Judith H. Mopsik | proc step, sas programmer, proc print, macro definition, sas programming, macro facility, proc freq | source | 6 | |
| 154 | The SAS/GRAPH ANNOTATE Facility | Anthony L. Friebel | Since its introduction in 1980, users have sought ways to create specialized graphic output with the SAS/GRAPH product. Some have been more successful than others. While creative manipulation of data can often produce the desired picture, a major criticism of the SAS/GRAPH product has been its inflexibility in dealing with descriptive annotation. This paper will explore new options for customizing standard SAS/GRAPH output, as well as new SAS/GRAPH applications |
dataset, proc step, sample | source | 8 |
| 155 | The Use of CATMOD for Repeated Measurement Analysis of Categorical Data | Gary G. Koch, William M. Stanish |
response variable, anova, predictor variable, sample, proc standard, repeated measures, proc calis, proc freq, main effect, proc step, maximum likelihood estimation | source | 10 | |
| 156 | The Use of SAS Software in New Product Introductions | Joyce D. Geier | proc step, monte carlo | source | 5 | |
| 157 | The Use of SAS Software in the 1DB | E.F. Mastal, G. W. Morrison, K.J. Notz, V.G. Trice |
The Integrated Data Base (IDB) Program prO vides official U.S. Department of Energy (DOE) data on spent fuel and radioactive waste inven tories, projections, and characteristics. This information is provided through the cooperative efforts of the IDB Program and DOE lead offices, lead sites, major programs, ,and glmerator sites. The program is entering its fifth year, and major accomplishments are summarized in three broad areas: (1) the annual inventory report. including ORIGEN2 applications and a Quality Assurance (QA) plan; (2) the summary data file and direct user access; and (3) data processing methodology and support to other programs. Plans for future work in these areas are out lined briefly, including increased utilization of personal computers. Some examples of spent fuel data are given in terms of projected quan tities for two growth scenarios, burnup and age profile of the existing inventory | dataset, data wrangling | source | 3 |
| 158 | The Use of the SAS System under VM/CMS in Teaching Introductory Statistics: A Model for Instructing First-Time Computer Users | Delwyn L. Harnisch, Carolyn J. Palmer |
This paper describes a project conducted at the University of Illino;s at Urba~a Champaign, wherein the SAS system under VM/CMS is used to complement and supplement mater; a 1 presented in an i nt roductory statistics course, Undergraduate and graduate students from a variety of academic departments are required to participate in weekly computer laboratory sessions and to complete homework assignments and a course project which require the understanding of statistical concepts and the use of appropriate SAS statements and procedures. This presentation will emphasize the importance of designing appropriate instructional materials, developing effective teaching practices. and providing an educational environment which will enhance the ability of first~time SAS users, first time computer users, and first-time statistics students to learn to use the SAS System under VM/CMS as a tool in performing calculations and solving problems presented in an introductory stat i 5t ics course | correlation, proc corr | source | 4 |
| 159 | THECIS-The Clinical Information System | Judith A. Burns, Leon T. Hairie, Thomas R. Hoffman |
crf, data validation rules, medical history, clinical trial, data science | source | 6 | |
| 160 | Transferring Data Between the SAS System and FOCUS | Edward P. Spotnitz | proc step, dataset, correlation | source | 3 | |
| 161 | Use of PROC FSEDIT to Generate SAS Code | C. David Hardison, Roger D. Muller |
SAS code, in general, has many structured features. This is f'specially true of SAS procedures which have a known number of options, procedural information statements, etc. Likewise, certain operations in the DATA STEP such as data set definitions and queries can also be structured. These structures lend themselves to fill-in-the-blank menus foe the naive user which, when peocessed, pass the user's input to E SAS code generatoe. Heretofore, these menus were developed using the SPF Dialog Manager and CLIST::; (or EXECs) with the eode being generated o~tside SAS. We have found that the SAS full scree~ editor available via PROe FSEDIT in SAS/FSP provides the same functionality as the Dialog Manager but with many additional benefits. The::;e benefiLs incluue Lhe fle:!dbility of placing the user directly in the SAS envi ronment |
proc step, data step | source | 4 |
| 162 | Use of SAS Software for Trend and Sequential Trend Test Analysis | Thomas P. Capizzi, Hina Mehta, Leonard Oppenheimer |
Experiments using a series of increasing doses of a compound, including the zero dose control are often used to study the effect of a test agent on the response of interest. The | proc step, proc glm, proc genmod, regression analysis | source | 6 |
| 163 | Use of SAS Software to Process and Analyze Animal Skin Tumor Response Data from a Data Acquisition System | R. Dana Laurie, Judy A. Stober, Freda Truman, Thomas Wessendarp |
SAS software is being used to check and summarize vast quantities of data. The data comes from computerized collection of data in an animal housing facility by a Hewlett Packard 9825 "Calculator". The data is collected weekly for a peTiod of one year and consists of loca tion, size and type of each skin tumor observ ed. In addition, study number, treatment number cage number, animal number, week of the study, operator ID, weight of the animal, and date of the current observation are obtained for each weekly data set. A typical study will contain 600 animals in 15 to 20 treatment groups. In order to monitor each study, two SAS software programs have been implemented. One (PROGRAM CHECK) helps assure that the data is correct and the other (PROGRAM SUMMARIZE) su~marizes the data in a manner useful to a toxicologist | proc step, proc freq, proc sort | source | 6 |
| 164 | Use of the SAS and SAS/FSP Products to Make the MSA/QPAC-Salary Package 'Online' | Hans D. de Roos | date variable, missing value | source | 2 | |
| 165 | User-written Device Drivers | Howard Houston | bufsize, char. | source | 6 | |
| 166 | Using A SAS Data Base | Carol Lambros | proc step, dataset | source | 2 | |
| 167 | Using New Features of the SAS System for Minicomputers | Stephen M. Beatrous, Bruce M. Tindall |
proc step, proc copy, libname statement, dataset, select, attrib statement | source | 6 | |
| 168 | Using New SAS Features in an Information Processing Course | Robert Carr, Karen Griffin, Robert S. Workman |
An introduction to SAS is taught as part of the sophomore level Information Processing course. In the Spring term of 1983 SASaZ was made available to us. The new SAS Procedures Tabulate, and Calendar made excellent additions to the SAS Procedures taught in the course | standard deviation, proc means | source | 2 |
| 169 | Using SAS Procedures to Examine Monthly Levels of Industrial Natural Gas Sales | John H. Herbert | regression analysis, proc princomp, condition index, proc calis | source | 4 | |
| 170 | Using SAS Software for Fatigue Life Prediction of Mechanical Springs | Robert L. Woell | proc nlin, sample | source | 5 | |
| 171 | Using SAS Software for Generalized System Testing and Verification | Claude Briggs, Katie A. Hubbell |
proc print, sas log, proc step, cpu time, proc freq | source | 4 | |
| 172 | Using SAS Software To Do Non-Parametric Statistics | Gerald R. Hobbs | In this article the use of SASe soft ~are to do nonparametric statistical analyses is explored. It is assumed that the reader is familiar with the software package and has some fami liarity ~ith parametric statistical | sample, proc step, anova, proc rank, degrees of freedom, null hypothesis, proc anova, response variable | source | 8 |
| 173 | Using SAS Software to Produce Summary Tables of GLM-Calculated Statistics | Margaret A. Grady | anova, least squares mean, title statement, cpu time, covariance, sample | source | 6 | |
| 174 | Using SAS Software with TSO/SPF to Monitor Network Availability | Anita Blitz, Gail Levine |
This paper will describe the development of a system to track problems related to various hardware components in a nationwide on-line health claims payment system. The Net~ork Availability Logging and Reporting System (NALARS) was developed to fulfill a need for accurate, complete information on vendor hardware problems. It was done in two stages. We will discuss the reasons for the two-part development method, the extensive error-checking, a high deqree of user-friendliness and the use or SASI in accomplishing these goals. NALARS is used by two distinct groups of people--network control operations staff and network management personnel. The needs of both groups and the solutions to their respective problems will be explained. To accomplish our goals of ease of use and assurance of accura 9Y , we employed many special features of SAStR. The use of these features ~ill be described as will the interface between SASlR) and TSO/SPF | trouble ticket, phase ii, phase i, conditional do loop, put statement | source | 6 |
| 175 | Using the SAS and SAS/GRAPH Products for Transportation Analysis | Gary F. Plazyk | proc step, proc gmap | source | 8 | |
| 176 | Using the SAS System to Perform Univariate and Cross-Spectral Analysis .. | John C. Brocklebank, David A. Dickey |
source | 7 | ||
| 177 | Using the SAS/GRAPH Device Driver for the IBM Personal Computer | Mike Kalt | source | 2 | ||
| 178 | Using the SAS/GRAPH Product for the Display of Environmental Matrices | James R. Carter, Robert A. Muenchen |
proc step, contour plot | source | 6 | |
| 179 | VAX Family--The Computing Strategy with a Future | Theresa Stokes | relational database, data dictionary | source | 2 | |
| 180 | Workload Characterization Using PROC FASTCL US | H. Pat Artis | Workload characterization is one of th~ most common techniques used in computer performance e'/aluation. This paper discusses how the FASTCLUS procedure can be used to analyze workloads. In particular, it discusses scaling and the treatment of outliers, how to effectively use the objective radius parameter, and the idiosyncrasies of the FROC FASTCLUS algorithm. 1.0 Introduction During the early 19705, Domenico Ferrari introduced the concept of workload characterization (FER72). The basic premise of the concept was that studies of compu~er systems could be facilitated by the identification of a number of "resource patterns" lhat could be used to statistically represent a systemls workload |
standard deviation, outlier, proc fastclus, proc step, cpu time | source | 8 |
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.