HOME QUICK SEARCH AUTHORS KEYWORDS HELP BUGS

SUGI 1991 Papers - Index

296 papers. Source: SAS Users Group International conference proceedings.

Back to Conf
go 1992
go 1990
# Title
×		 Authors
×		 Abstract
×		 Keywords
×		 Extras Pages
1 "Rumors of My Death Have Been Greatly Exaggerated ": Mark Twain Discovers SAS/SHARE Software in Version 6 Bill Brideson This paper reviews the progress of SASISHARE" software since Release 5.18 of the SAS· System, induding demonstrations at SUGI conferences and the current experimental release program. It highlights new features that have been added to the software, and it sketches SAS Institute's plans for the software for the near future source 5
2 1990 Census Mail Response and Related Variables: An Examination at the County Level Using Regression Analysis Jay K. Keller,
Quentin Newhouse Jr.		 predictor variable, correlation, regression analysis, response variable, dif function source 5
3 A Blueprint for Interactive PC-Based Applications Development Aiman M. Zeid proc step, catalog, sas output, macro variable, proc printto, macro definition, options statement, libname statement, sas base source 10
4 A Clinical Data Review System to Facilitate Pharmaceutical Research Martin J. Rosenberg In an effort to better manage the enormous volume of information and to reduce the length of time required to introduce new drugs in the Untted States, the U.S. Food and Drug Administration and the pharmaceutical industry have sought to facilttate the drug development and NDA review processes through innovative uses of computer technology. One such effort is the clinical data review system, sometimes also known as a medical review system. Clinical data review systems permtt monttors, CRA's, managers, and other members of the clinical staff to monttor the progress and outcome of ongoing clinical trials. They can also serve as an important step in the development of a computer assisted NDA or CANDA. These concepts are depicted through a demonstration of the CUNACCESS TM clinical data review system. Originally developed under Version 5.18 of SAS, CUNACCESS has been substantially enhanced under Version 6 data review, crf, clinical data, proc means, pharmaceutical industry, data dictionary, key, clinical research source 9
5 A CMS-based, Multi-User Planning System for Clinical Data using SAS/AF and SAS/FSP Software Thomas E. Schneider,
Mary Catherine Krug		 proc step, clinical trial, bar chart, vertical bar chart, sas log, sample source 8
6 A Comparison of DATA Step Performances in Releases 6.06 and 5.18 Jimmy A. DeFoor where statement, dataset, libname statement, output, bufsize, if-then-else, proc print, proc step, sas library, subsetting if source 6
7 A Comparison of RS/1 and SAS Software in Meeting Information Systems Development Needs .. . Loren L. Heun,
Michael S. Lajiness		 data science, statistical software source 8
8 A Data Entry Application in Clinical Research Using Quality Assurance Tools Miriam W. Rosenblatt Although data entry and quality assurance systems have been well described, practical applications synthesizing such methods are not always obvious to the user. This paper describes an inexpensive data entry application using quality assurance tools for entering multiple data forms collected at different time points. Data entry was conducted in interactive mode on a mainframe (OS/MVS) using SAS/FSP*, Version 5.18; data management was conducted in batch mode using Base SAS, Version 5.18 under ISPF. Simple macros were developed and maintained in batch files for automatic sorting of databases, comparing databases for' cleanup,' and updating master files. Double entry was used to assure data quality. This application is targeted for small academic research groups, requiring computer-intensive resources for entering, managing and analyzing data, especially for larger studies quality assurance, proc step, clinical research, data cleaning source 6
9 A Debugging Facility for SAS Macro Systems David S. Frankel,
Mark A. Kochanski		 This debugging facility makes tt easy to trace the execution of one or more SAS macros within a SAS job. Vou can control printing start- and end-banners that identify lIacros, including revisIon date and support group; checking the values of macro parameters; printing notes following PROC and OATAsteps; printing a stmple or detailed source-code trace; and, deletion of temporary SAS data sets. Macros not to be traced execute Has isll with settings determi-ned by their Invocations and definitions. The facility consists of two SAS macros - %DEBUG and ~ DEBUG. An end-user or developer call. %DEBUG from a SAS job to set the debugging options to be used for subsequent macro calls. Each SAS macro call s ~ DEBUG to print banners, set status variables, delete.temporary SAS data sets, and replace statements like OPTIONS NOTES. ~ DEBUG determines the environment for each micro by arbitrating between .lIas isrjo settings and %DEBUG user requests macro definition, proc step, macro variable, options statement, macro parameter, macro call source 6
10 A Factor Analysis of Employee Attitudes about the Internal Operational and Administrative Characteristics of a City Government Anthony C. Waclawski It is a well established fact that there is an inextricable link between an individual's perception of the internal operational and administrative characteristics of hiS/her place of work and the quality of products produced and/or service delivered to the customer. Decision makers In both the public and private sectors are beginning to widely acknowledge this relationship and are beginning to develop strategies to address identified areas of concern. Ouring early spring of 1990 a scientific survey of employees was conducted in order to measure attitudes about the internal operational and administrative characteristics of acitygovernment. This paper describes the results of extensive statistical analyses of the survey data. It demonstrates how the power and flexibility of SASe can be used to factor analyze a (212 observations x 41 variables) data matrix in order to isolate specific factors employees view as being important to the internal administration of an organization factor analysis, anova, sample, factor loading, proc freq, numeric variable, correlation, central tendency, data step source 8
11 A Flexible User Interface Developed Using SAS/AF and SAS/GRAPH Software Jane P. Loftis,
Teresa L. James		 Oak Ridge National Laboratory (ORNL) was asked by the MilHary Traffic Management Command's PtifSOnai Property Directorate (MTPP) to design and prototype a decision support system to quickly and easily locate and display data to assist them In their task of evaluating and monHoring the Personal Property Program. This paper describes the user interiaoe developed In SAS/AF' and SAS/GRAPH" that supports this system and explains how the SAs" user Interiaoe wor1<s In oonjunclion wHh an ORACLE" Interfaoe. It also discusses the goals of the SAS user Interface, describes the screens and pop-up windows developed for the interface, and explains how these screens achieve the system goals bar chart, line chart source 6
12 A Macro for Analyzing Insulin or Glucose Concentration Time Curves Alfred J. Anderson proc step, proc genmod source 6
13 A Macro for Multivariate Statistical Process Control Meng-Koon Chua The usefulness of multivariate statistical process control techniques has been received a tremendous attention. However, the lack of computer tools presents one of the maior obstacles to the acceptance of these techniques. A macro is thus developed to provide two control charts: the multivariate exponentially weighted moving average control chart and the log lSI control chart. These control charts constitute a multivariate analog of univariate control charts and are useful in detecting small process mean shifts and process dispersions respectively. Some case studies under linear mean trend shift and covariance expansion situations are given to demonstrate the performance of the control charts sample, covariance, normal distribution source 5
14 A Macro For Principal Component Logistic Regression Brian D. Marx Marx and Smith (1989 and 1990.b) described the detriments of an ilkonditioned information matrix to lo gistic regression. Estimation and prediction problems, siJn.. ilar to those resulting from multicollinearity in standard multiple regression problems, can also exist in the logistic framework with ill-conditioning. In particular, large vari ances can occur for the maximum likelihood parameter es. timates, as well &8 for predicted values in certain regions of the variable space, among other prob~ma. A3 an alter native to maximum likelihood estimation, Schaefer (1986) and Marx and Smith (1990a) presented asymptotically bi ased one-step and iterative principal component parameter estimators, respectively, for the logistic setting. Principal component techniques can be particularly useful when vari able deletion is not a reasonable alternative and theoretical models are well defined proc princomp, logistic distribution, regression analysis, logistic regression, maximum likelihood estimation, likelihood function, predictor variable, proc calis, proc logistic, proc step source 6
15 A Macro Program for Generating Customized Reports from User-Specified Column Variables and Page-Sort Variables Jimmy A. DeFoor macro variable, &variable, variable, call symput, if-then-else, %if-%then-%else, numeric variable, quality assurance, proc step, macro loop source 7
16 A Macro to Create a SAS Data Set Containing Commonly Reported Statistics for the Analysis of Multiple 2 x 2 Tables Suzanne Edwards,
John Henry King,
Gary G. Koch		 confidence interval, odds ratio, proc freq, proc step, sample, fisher exact test, dataset, exact test, proc calis, response variable source 4
17 A Maintenance Tool for Reporting on Release 6.06 SAS/AF Applications Chris Williams flat file, catalog entry source 6
18 A Procedure for Deciding Which Treatments to Accept in a Screening Study Involving Pseudo-Bernoulli Data Ralph A. DeMasi,
Dana Quade		 sample, contour plot, sample size, standard deviation, power analysis, stage 1, null hypothesis, stage 2, binomial distribution, proc step source 7
19 A Program for Outputting P Values and Confidence Intervals in an Analysis of Variance without the Use of the GLM or PR/NTTO Procedures Susan Foley Anton,
Joseph J. Andary		 anova, proc calis, change from baseline, p-value, proc step, response variable, regression analysis, proc glm, binary variable, proc genmod, confidence interval, proc printto source 6
20 A Program to Compute Odds Ratios and Confidence Intervals from LOGISTIC Output Eric Yu I. Wang,
Aili Gong,
David P. MacKinnon		 The odds ratio and its 95% Confidence Interval are widely used in epidemiology and clinical trial research. The purpose of this paper is to describe a program to automatically compute these statistics with the SASjSTAT LOGISTIC procedure and the SAS MACRO product. The program can be used for both binary and ordinal response measures with one or many independent variables. The use of the program is demonstrated in two examples from prevention research logistic distribution, odds ratio, estimate, proc step, predictor variable, regression analysis, response variable, confidence interval, logistic regression, proc logistic, standard deviation, macro definition, proc calis, proc sort source 4
21 A Reasonably Comprehensive Test of SAS Programming Skills Standish Sibley Not infrequendy. a programmer is chosen for a SAS programming position after what may amount to an inappropriate interview. Per haps congeniality is achieved and applicant and interviewer per sonalities mesh weH, potentially resulting in a job offer for the wrong reasons. Perhaps references weren'tadequatelycheckedorhonestly appraised; perhaps the applicant did not-for a variety of reasons supply examples ofcode demonstrating his or her unaided skill level. In these cases, the hiring manager has litUe Kleaofwhattheapplicant is truly capable of, and offers of employment or salary level may be based on entirely inappropriate aiteria This paper has associated with it a variety of test items designed to qualify an applk:ant. containing questions and problems revolving around real-wor1d applications. The items range enough in difficuhy to identify the beginner or tax the expert. and so may be used to help make the right hiring decision proc step, macro variable, numeric variable, run, sas programming, sample, macro facility, date variable source 5
22 A Roadmap to Macro Facility Error Messages and Debugging Susan M. O'Connor sas log, macro facility, macro variable, macro definition, error message, catalog, macro quoting function, macro function, error, macro loop, macro put, &variable, stored macro source
23 A SAS Macro for Assessing Differential Drug Effect Terrance L. Fox,
Bruce Ekholm		 Regression to the mean can create an incorrect impression of the relationship between drug effect and baseline measurement. We propose a SAS Macro for adjusting for the regression effect and testing for differential drug effect. The macro is applied to blood pressure data regression analysis, degrees of freedom source 4
24 A SAS Macro for the Robust Analysis of Repeated Measures Data Joe DiGennaro,
William J. Huster		 repeated measures, crossover study, binary variable, anova source 6
25 A SAS Procedure for Outlier Detection in Bioavailability/ Bioequivalence Studies Jyh-Shiun Lin,
Shein-Chung Chow,
Siu-Keung Tse		 In the pharmaceutical industry, a crossover experiment is often conducted to assess bioequivalence between a test formulation and a reference formulation of a drug product.Two formulations are considered bioequivalence if the ratio of means (test/reference) of the pharmacokinetic parameters of interest such as area under the blood concentration-time curve (AUe) and maximum blood concentration (Cmax) is within some reasonable limits (say, (80%, 120%" with 90% assurance. A bioavailability analysis that includes possible outlying values may affect the decision on bioequivalence. Chow and Tse (1990, statistics in Medicine, 9, 549-558) proposed two test procedures the likelihood distance (LD) and th~ estimates distance (ED) for detection of an outlying subject. In this paper, a SAS procedure for the LD and ED test procedures is developed using the SAS Macro languages. An example is presented to demonstrate the use of the SAS procedure likelihood function, confidence interval, macro facility, macro definition, outlier source 5
26 A SAS Program for Calculating and Displaying Regression Lines in Two-Way ANCOVA Models with Unequal Slopes Joseph J. Andary When the assumption of parallel slopes is tested in an ANCOVA, the regression slopes (the interac tions with the covariate) compared by the SASe system must be correctly calculated and graphed so that the nature of any heterogeneity in slopes can be investigated. In a one-way ANCOVA, this is straight forward since the least -squares method can be used and the resulting regression lines could be graphed by using PROC GPLOT. However, in a two-way (or higher order) unbalanced deSign, only the slopes in each cell of the design (three way interaction with covariate) can be displayed by a least-squares regression analysis, sample, sample size, proc step, ancova, proc gplot, covariance, macro definition source 5
27 A SAS Program for Choosing Random Unblocked Designs Balanced for Correlation between Neighbors I.J. Good,
Eric P. Smith,
David D. Morris		 David D. Morris, Abbott Laboratories . with treatment tIl as plot number q + 1. Thedeslgn is then balanced for pairs, but there would be one extra treatment t 1" The purpose of the balancing is only to remove some of _ the effect of correlation between adjacent pairs, and the analysis of a completed experiment is to be carried out without correlation, sample, proc freq, macro definition source 5
28 A SAS Program for Seasonal Kendall Trend Analysis of Monthly Water Quality Data Sungsue Rheem,
Golde I. Holtzman		 A SAS program that automates the seasonal Kendall water quality trend-analysis and flow-adjustment procedures of Hirch et al. (1982) is presented. The input is monthly observations of a water-quality variable and contemporaneous stream-flow meas urements over a perio~ of several years. The program consists of three parts. Part 1 peIfonns the unadjusted Seasonal Kendall procedures, producing SUllllIlaIY statistics, a plot of the water quality variable against time, overall Tau and the P-value of the test for trend, monthly Tau values, and the Seasonal Kendall es timate of slope with corresponding confidence intervals. Part 2 is a maximum-R2 model-selection procedure for regression of the water-quality variable on a function of flow. The output of part 2 includes a list of the candidate models, summary statistics, a plot of the observed and predicted water-quality values against flow, and a statistical description of the selected model proc step, proc print, regression analysis, confidence interval, proc means, macro definition, correlation, feature selection, trend analysis source 6
29 A SAS/AF Application for Developing Project Schedules Using SAS/FSP and SAS/GRAPH Software Sharon M. Haney,
Laurie T. Elstun,
Kelly K. Antony		 This paper summarizes a SAS/AF menu driven application developed for project management. In the pharmaceutical industry. project management is a vital. yet time consuming task. At Marion Merrell Dow Research and Development we must be aware of the status of many clinical trials and be able to summarize the time a trial has spent in a certain phase of development. This project schedule system facilitates project management by creating and storing pertinent dates associated with the phases of development of £rojects. especially clinical trials. in a SAS dataset using SAS/FSP. The progress of the projects is graphically displayed over a specified p'eriod of time using SAS/GRAPH. This system enables analysts. managers. consultants. clinical research associates and M.Do's to manage their projects. as well as their work load. efficiently and effectively clinical trial, proc contents source 6
30 A Sophisticated Algorithm for Annotating SAS Polygons Rich Priem,
Anita Clark		 Exxon is in the process of converting all of their geoscience mapping applications to SAS Graphics. A significant portion of the mapping needs include representation of land ownership. Traditionally, drafters have maintained complex hand-posted linens depicting the company's lease holding position. Due to multiple ownership, and the multitude of pertinent lease information, annotation standards have evolved to great complexity. To automate the digital mapping of lease polygons, a sophisticated annotation algorithm was required. This paper presents the specifications, design, and implementation of a robust SAS Macro to solve this annotation challenge. The final product is a set of annotations which are aligned to a dominant trend and strategically positioned to fit multiple lines of text within both simple and complex polygon' outlines macro definition, dataset source 5
31 A System for Pinpointing Areas. Needing Process improvement Bruce T. Blatzer proc step, proc print source 3
32 A Tale of Two Releases: Benchmarking the Performance of the SAS System Release 6.06 against Release 5.18 Michael A. Raithel dataset, cpu time, bufno, bufsize, libname statement, batch job, index creation, sas library, proc step source 10
33 A Tool to Create a Cross-Reference Listing of Variables and Literals in a SAS Program Kathy Wilhoite proc step, macro variable source 6
34 A Window-based Graphics Toolkit Using SAS/IML Software Charles B. Hallahan As a matrix programming language, SAS/IML is a natural language for statistical and mathematical applications. However, IML can be used for much more. With commands available for dataset manipulation, window control, high resolution graphics, module definition and dynamic creation of I ML statements, complex, yet easy to use, programs can be written in the IML language. This tutorial will demonstrate the versatility of IML by developing a menu-driven graphics toolkit called GTOOLS. Some general guidelines will be given for doing similar projects dataset, character variable, conditional do loop source 9
35 About JMP 2.0 John P. Sall,
Katherine Ng,
Michael Hecht,
Ann Lehman		 confidence interval, sample size, sample, dataset, proc corresp, f-test, proc nlin, proc genmod, likelihood function, regression analysis source 6
36 ALMA: A Library Management Information System Using SAS/AF and SAS/FSP Software on a DOS NOVELL PC Network Carmen J. Vincent,
Linda Gowen		 This paper describes ALMA. an information system we developed to manage a departmental reference and lending library at Weslat. ALMA is programmed in the SAS/AF and SAS/FSP systeIWI and runs under DOS 3.31 in a PC network environment under Novell NetWare 386. When we first began to develop the library system. we did not know how to use the SAS/APE and SASIFSP. systems and the PC network: was in a highly experimental test phase, serving fewer than 100 of the SOO planned users. Our decision to implement the system using these products was a way of learning about the SASIAF and SASIFSP systems, and SASe software in a network environment. The ALMA system was also developed to be a demonstration program. This paper describes the design and functions of the library system, and presents some of the teaching materials we developed for presenting the system and the techniques used to create it. We have also included a description of the network, files proc step, dataset, symbol table source 5
37 An Approach: Meeting the Ad Hoc Reporting Needs of Clients Jewel Fries source 4
38 An Insider's Sneak Preview of the Next DATA Step Release Jeffrey A. Polzin modify statement, dataset, output, proc step, do loop, master data management, composite key, sysrc, select, by group processing, set statement, leave, proc print, proc sort, sample, macro facility, rename statement, update statement, automatic variable, continue, key= source 7
39 An Introduction to Features of the C Language Gary H. Merrill This paper is a brief introduction to the most signifrcant features of the C programming language and is intended as a quick introduc tion to that language for those who are already fam~iar with some high-le'llel programming language. It is not intended to be a complete description of C, but rather to give the flavor of the language and serve as a basis for further discussion in the question and answer session numeric constant, do loop source 6
40 An Introduction to SAS Functions Ronald P. Cody character variable, sas function, put statement, sysdate9, input statement, date variable, datepart, lag function, proc step, numeric functions source 6
41 An Introduction to the FSEDIT Procedure Marge Scerbo,
Steven H. Feder		 proc step, dataset, where statement source 4
42 An Introduction to the SAS Macro Language Gary L. Katsanis macro variable, proc print, proc step, macro facility, &variable, macro function, macro definition, macro loop, dataset, macro parameter, %if-%then-%else, if-then-else source 7
43 Analysis of 2 x 2 Cross-over Trials Using an Integrated Macro Package Harvie Chang,
Tamara R. Strang		 Two by two cross·over designs are frequently used in clinical trials. Jones and Kenward (1989) have outlined a framework for analyzing 2 x 2 cross·over trials. The framework includes techniques which can be used to quantitatively and visually assess the association between the treatments and the response of interest. We have developed a macro package, using SAS" procedures, to analyze data from a 2 x 2 cross·over design according to the framework. Most of the techniques implemented in our macro package are commonly used to analyze cross·over data. These techniques are sUbject profiles plots, a sequences by-periods plot, an analysis of variance table, residuals plots, and a confidence interval. The package also includes a less commonly used, but very powerful, plot which allows one to visually assess the carry-over and treatment effects treatment effect, anova, sample, outlier, null hypothesis, clinical trial source 6
44 Analysis of Designs with Detached Control Groups Robert M. Hamer dataset, proc step, main effect, proc glm, response variable, proc genmod, anova source 7
45 Analysis of Paired Categorical Data Using McNemar's Test and Bowker's Test Jaye Lynne Thompson The SAS System has long been the standard for statistical analyses, yet we often omit valid sample, macro definition source 7
46 Analysis of Sparse Repeated Categorical Measurement Data Alan Agresti,
Stuart Lipsitz,
Joseph B. Lang		 The feasibility of maximum likelihood (ML) analyses of marginal distributions of repeated categorical measurement data diminishes as the numbers of response occasions and response categories increases. This article describes alternative approaehes that are much more feasible. To estimate model parameters, we recommend a "pseudo ML" approach that treats repeated responses as independent and uses a jackknife to estimate the covariance matrix of tbose estimates. Tests of hypotheses about response distributions (e.g., marginal homogeneity) use Wald statistics or adapted score statistics from the independent-samples case. We illustrate these analyses with a seven-dimensional table having 78,125 cells. Simulation results show no substantive loss of efficiency from using pseudo ML estimates covariance, sample, likelihood function, logistic regression, missing value, proc calis, proc freq, multinomial distribution, proc step, missing completely at random, missing at random, proc means, maximum likelihood estimation source 9
47 Analyzing a Survey: A Bunch of Tips for a Mess of Questions William G. LeBlanc People who desire to analyze surveys must be compulsive, retentive, suspicious, manipulative, miserly. paranoid, and last but not least. lazy. These characteristiCS can be learned by those who are not fortunate enough to have been born with them. With the addition of some good programming habits and a few tricks, anyone can do a decent job of analyzing a survey proc step, input statement, numeric variable, proc format, character variable, missing value, proc freq, format statement, length statement, dataset, label statement, proc means source 4
48 Analyzing Product Test Data in a Relational Database Using SAS Software John L. Orman SAS software is being used to analyze product test data stored in an INGRES .... relational database. The database has been implemented at Allied-Signal in Kansas City on a Digital Equipment Corporation (DEC"') VAX'''' computer. The INGRES application development has been a joint project between Sandia National Laboratories and Allied-Signal. Application screens have been developed so that the user can query the database for selected data. Fourth generation language procedures are used to retrieve all data requested. FORTRAN and VAX/VMS- DCL (DIGITAL- Command Language) procedures are invoked from the application to create SAS data sets and dynamically build SAS programs that are executed to build custom reports or graphically display the retrieved test data along with control and specification limits. A retrieval screen has also been developed which invokes SAS software to calculate the mean and standard deviation of the retrieved data dataset, flat file, standard deviation, relational database, sample, proc sql source 6
49 Analyzing Transmission Conductors under Various Meteorology Conditions Using Macros and Graphics Patricia Gold,
HelenAnn W. Volpe		 This study determines the relationship between actual and theoretical wind loads on transmission conductors. The purpose of the study is to improve Bonneville Power Administration transmission line design proc step, macro variable source 3
50 Automated Generation of Statistical Tables: Integration of Structured Programming and SAS Macros Lee-Hwa Jane,
Maureen McCormick		 dataset, macro call, proc means, proc freq, proc step, sample, sample size source 5
51 Being Dynamic Warren E. Stinson,
Phil Fisher		 data step, proc step, proc format source 4
52 Best Presentation of Data-Color Patrice Dor proc gmap, proc step source 12
53 Best Presentation of Data-Monochrome Paul W. Fingerman proc step, dataset source 8
54 Box and Whiskers Revisited - Using SAS/GRAPH Software to Produce Box and Whiskers Plots Hedy Weissinger It is often useful to look at data using basic summary statistics such as mean, median, minimum, maximum. and quartile values. A box and whiskers plot displays this information. giving an indication not only of the range of the data. but also an indication of the distribution of the data. It is a useful tool for exploratory data analysis. This paper describes a series of macros which produce these plots. The macros use ANNOTATE and GPLOT to produce a box and whiskers plot and can run under SAS version 5 or version 6 interquartile range, box plot, outlier, proc univariate, by group processing, macro call, proc step source 4
55 Building a VM Print Accounting System with Base SAS Software Orion V. Hanger flat file, extract, dataset, libname statement, sample source 7
56 Building and Submitting Jobs Using TSO and the SAS System Jody J. Fromer batch job, sample source 2
57 But I Don't Want to Learn the SAS System, System 2000 Software, Communications, or Timesharing to Get My Report Laurie Burch This paper describes EasyBCS, a microcomputer program designed to totally insulate the user from learning any of the multitude of products required to create, submit, execute and retrieve a job on a mainframe computer. EasyBCS, aided by extensive tables describing the mainframe database, walks the user through the creation of his job with "point-and shoot" menu selections. When completed, EasyBCS will automatically translate the user's request into SAse or System 2000· code and submit it to the mainframe for execution. When the job has completed. the user may either browse the output on the mainframe through EasyBCS, send the output to a selected printer, or download the output to the microcomputer. The user interacts only with EasyBCS. EasyBCS does all the necessary interaction with all the required software packages source 5
58 Coding A User Application into the Executive Information System of SAS/ASSIST Software Linda P. Atkinson proc step, dataset source 8
59 Comparing File Size to Free Disk Space When Copying DOS Files from within SAS/AF Applications in Release 6.03 Craig Bickel source 2
60 Comparison of Mainframe Hardcopy Devices Compatible with SAS/GRAPH Software LeRoy Bessler sample, cpu time source 4
61 Comparison of SAS/GRAPH Software with Other Mainframe Graphics Software LeRoy Bessler batch processing, proc step source 2
62 Competing Risk Multi-Spell Survival Analysis: The CTM Procedure Dan Steinberg competing risks, likelihood function, proc phreg, hazard function source 6
63 Computer Performance Evaluation Using SAS/CPE Software under VMS: A Case Study Virginia M. Dineley System performance probtems are often difficult to track. and diagnose. This paper uses a case history approach to serve as an example of the use Release 2.0 of SASlCPE software to diagnose some specific real wood performance problems in a VMS"" environment. The iterative nature of tuning adjustments is discussed along with exa~les of how SASlCPE Software can be used to indicate the effectiveness of tuning adjustments cpu time, raw data source 11
64 Confidence Intervals for Contrasts in the Unbalanced Analysis of Variance Using SAS/IML Software John Ilgenfritz,
Anand K. Seth		 When performing an analysis of variance, one often requires the use of confidence intervals in the reporting of statistical results. The SAS· GLM procedure does not give confidence intervals for contrasts as part of its output. Through the use of SAS Interactive Matrix Language (IML), a program has been written to calculate these confidence intervals at any desired alpha level. I. Introduction Use of confidence intervals in the reporting of statistical confidence interval, proc genmod, proc glm, anova, proc calis, degrees of freedom, fixed effects, proc mixed source 6
65 Contrast Generating Macros for the GLM Procedure John Pesek In this paper we will describe some macros that we have written to generate contrasts for the GLM proce dure. The CONTRAST statement is a powerful tool for analyzing experiments using planned contrasts. It is clear that it would be more powerful if it were possible to indicate certain common contrasts concisely instead of writing out the coefficients. In fact, this facility is provided in the REPEATED statement. With it one can request any of several sets of contrasts including orthog onal polynomial contrasts for trend analysis. It is easy to see most contrasts involve just 5 concepts which are: 1. One level compared to another (simple comparison). 2. The average of one set of levels compared to the average of another set of levels (complex compari son). 3. Orthogonal polynomial contrasts. 4. Comparisons restricted to a level of another factor (or factors). 5. Interactions of one or more comparisons with an other proc genmod, proc glm, macro variable, proc step source 6
66 Conversion of a SAS/AF Graphics Reporting System from Release 5.18 to Release 6.06 in the TSO Environment Darlene M. Gerber In this paper I'll descnbe the process of converting a Release 5.18 SASIAF" Oraphics Reporting System to Release 6.06. I'll discus. enhancements such as DIRLIST which displays a list of datasets returning the user's selections. I'll also discuss the problems I encountered converting SASJAF cata logs. printing graphs. and gaining experience with Screen Control LanguageiSCLI. This system lVILS developed on a Compaq DESKPRO 2868 connected to an IBM Model 3090'. All output was printed by a DEC LN03 Plus" printer using the TEKLNOI Device Driver proc step, libname statement source 5
67 Conversion to Version 6 of the SAS System: Experiences and Advice from a Management Information Systems Department Jerry J. Hosking,
Cliff Schell		 Discussed here are experiences and advice gathered during initial conversion efforts to Version 6 of the SAS· System by the Manage ment Information Systems Department at SAS Institute. Our organi zational approach is explained and specific details of changes required and recommended for the conversion process are pro vided. Selected enhancements to applications using new function ality available through Version 6 are highlighted libname statement, batch processing, options statement, proc step, data cleaning source 5
68 Converting Applications from Release 5.18 to 6.06 Melissa Kerner This paper presents eKamples of converting commands and methods used in Release 5.18 to those supported in Release 6.06. Emphasis will be placed on equating commands in both versions and adapting code to utilize the power of the new versionr ~xamples will include: command cross-reference list Data communication between . SASj5CL programs and SAS datasets * Helpful new S~S/Display Hanaqer commands Eglliyalegt cmueands 1.) Assigning DDnames to fi1es and libraries. Version 5.1B TSO ALLOC F(DDname) DA('SAs library') dataset, where statement, proc step source 4
69 Creating a User-Friendly Statistical Processing System in a Batch Environment Julia L. Bienias dataset, cpu time, proc step source 6
70 Creating Course Notes from an Annotated SAS Program Karen G. Malley A nonstandard approach to the prgparation of course notes for an introductory-level SAS course Is described. The use of a thoroughly-commented SAS log provides a m9Chanism for easy updating as new features of SAS are 8vaUable, and appeals to students with diverse backgrounds since there can be much to look at and learn from in the SAS log sas log, proc step, proc print, sysdate9, date variable, length statement, put statement, run, batch job source 4
71 Creating Dot Diagrams Karen G. Malley The Dot Diagram macro is used to create a type of frequency plot called a dot diagram, similar in concept to a histogram. Each dot on the picture represents an observation of a numeric response variable. thus show:ng the general shape of the distribution. The horizontal axis of the diagram represents discrete groups into which each observation is categorized (see Figure 1). Dot diagrams are suited to a small number of groups, The macro will print one dot per observation with non-missing group and response values. Dots are centered over the group lahel. If you have a large number of groups (more than ten), or a large number of observations (more than 1000), a histogram is recommended, although this macro will still work. PROC CHART or PROC GCHART will produce histograms. Two pictures may be produced: one from PROC GPLOT suitable for use in publications, and one from PROC PLOT to give the user a rough idea of the distribution by group. No class variable, proc step, macro call, response variable, proc freq, macro loop, macro definition, macro conditional, histogram source 4
72 Customized In-House Training with a SAS Trainer's Kit Marcia S. Murto data science, proc step source 3
73 Data Analysis Applications for the Semiconductor Industry Leslie Fowler,
Alan Yost,
Jim Chalmers		 Effectively analyzing the enonnous volume of data collected in the semiconductor industry requires acWanced end-user applica tion development Data is coUected throughout the 100 or more manufacturing process steps and during eleebical tests per formed after processing is complete. These data need to be analyzed .od conelatad for ~ in producing higher per fonnance devices. The SAS System -was used to create menu-driven applications for the end-users. Oatalog is an on-Une process control data entry and analysis system. This system feabJres a hierarchical organization for collecting data in various manufacturing areas which contain multiple-variable charts, histograms, normal and shewhart type charl!;. The 'Engineering Data Analysis System (EDAS) is a menu driven system containing procedures to aid engineers in analyz ing data. This system provides the tools 10 manipulaIB, plot, statistically analyze, summarize, trend and correlate electrical test data and manufacbJring process data histogram, anova source 5
74 Database Design Options Using Release 6.06 of the SAS System Tim Feetham,
Sally Witt,
Lauri Vance		 key, foreign key, normalization source 3
75 Database Design Using Compiler Theory Techniques Wade A. Pedlow With databases becoming larger and more complex, sometimes it is advantageous to create database calls which are a bit more sophisticated ~an t~itional cal!s. Database Design using Compiler Theory Techniques IS an overview of a project which involved the ~ation of a la~ database w~ich could be transparentty accessed In an~ order, not Just the key, without destroying response time. The compiler theory techniques utilized are: Finite State Machines Parsin~ Production Rules, Tree Structures, Look-up Tables, and Recursion thrown in for some fun where statement, dataset source 4
76 Database Update with Incremental Audit Trail and Rollback Using the FSEDIT Procedure and Screen Control Language William R. Fehlner A ecilica1 component 01 many database update appli<:adons 1.lI1e ability 10 maintain a comple19 audit lrail, even when update Is inter· rupted. This paper desaibes the implementation of an itJOml"OOntal audit trail within the FSEOIT Procedure using Screen Control lan guage (SCL). This appliOatlon ",no without change on _ platfonns under Release 6.06d!he $AS-System. Audit cIa1a ..... as CUfJent as each updated observation, allowing fOOQ't,i'Cry from UI'}-o expected interruptions to be comp4ele. Auditdala is stomd in a form which makes possible automatic rollback 10 earlier &lales 01 the dalaba.e. The SCL """" that genera"," audit data Is writWn 1D ba independent 01 other SCl funClions thai may ba ntqUired by !he FSEDITappftcatlon.Oependenooonlhocla1astructureoitheediled dala view or data file Is limited 1<1 a omaII £&gmw 01 SCl master data management, audit trail, missing value, proc step source 4
77 Database Validation: A Random-Sample Audit Procedure Bruce Wentworth,
James M. Harper,
Barbara A. Costantini,
David A. Bennett		 sample, sample size, crf source 4
78 DB2 Performance Management Donald T. Gantz,
James M. Thompson		 This paper describes analysis done on measurements of a DB2 production workload at a Fortune 500 company. DB2 Is IBM's strategic database management system. Several sections of the company's computer and information systems staff have a need for DB2 performance information. The authors have developed and piloted a technique for relating the design structure of DB2 plans to their performance. SAS· Institute's products play a major supporting role in executing this technique. SAS programming, procedures, report writing, and graphing are used to augment and exploit a computationally intense classification tree software product sample, cpu time, elapsed time, regression analysis source 10
79 Decision Tables as SAS Analysis and Programming Tools Anne M. Asher,
Todd Peer,
Elizabeth F. Henrickson		 letters of credit, sample, sysdate9, proc step source 6
80 Design Considerations for Optimizing File Processing in an Integrated Data Management and Analysts System Michael R. Wrona,
Norman A. Constantine,
Mark P. Branagh		 proc step, dataset source 5
81 Designing a SAS Software Based System to Track Multistage Data Processing Tasks for Multiple Databases Laurie A. Cameron This paper describes a $AS software based system designed to track multistage data processing for a large number of data sets in a recent project undertaken by Westal, Inc. A system was required because a number of analysts were involved in the project. fts primary purpose was to verify that aU prerequisite prooessing for a given stage was complete before processing that stage. Each stage had to be manually approved before proceeding to the next stage. M. any stage, the approved status of some prior stage could be reversed, requiring that it and all processing for which it was a prerequisite be resubmitted. The system was sufficiently flexible to allow for changes in processes that could not be fully specified before examining the data data wrangling, raking source 2
82 Developing a Barcoded Tool Tracking and Checkout System using SAS/AF Version 6.06 Quinton J. Ritz,
Dawn M. Thomas		 The Perry Nuclear Power Plant located in North Perry. Ohio. is required by Federal Law to keep records and data on the usage of calibrated test equipment used in the maintenance and testing of Nuclear Power Plants. The "Measuring and Test Equipment Tracking System" (METS) was devel oped to provide this traceability using SASI~ 6.06 software and barcode technology. The auto mated collection of data will provide quick and ef ficient information as to the status of test equipment dataset, proc step source 2
83 Developing a Data Entry and Retrieval Application Using SAS/AF and SAS/FSP Software Lisa Ann Horwitz The features of Release 6.06 SASI AP and SAS/FSP" software pro vide the tools to deveiop data entry and retrieval app!ications that perform extensive field and cross-fietd validation, referencing and tabie lookup, and querying. This paper looks at some of the ways Screen Control Language (Sel) functions, selection lists, and extended tables can be used to perform these tasks dataset, data cleaning source 4
84 Developing a Generic Application to Prepare Mailing Lists Preston A. McCoy macro call, sas base source 4
85 Developing a SAS/AF System for Statistical Process Control (SPC) Applications Melissa A. Durfee Utilizing SAS/AF menu capabilities, a customized systeIll for Statistical Process Control (SPC) applications and basic statistical analY~A:s may be developed. This system incorporates SAs/,gp for constructing Shewhart control charts, SAS/FSP f~{ displaying and storing control limits, SAS/GRAPH for constructing multi-vari charts and control charts with high-resolution graphics, and base SAS TM for statistical analyses and comparisons. Since a menu-based system is user friendly, the usage of SPC, including its problem-solving techniques, is facilitated proc step, dataset source 6
86 Developing Statistical Application Workshops with SAS Software on a PC Priscilla Hancock proc step, anova, dataset, regression analysis, proc glm, proc genmod source 2
87 Differences and Interrelationships Between the SQL Procedure and SAS/ACCESS Software Mark Stranieri The purpose of this paper is to compare and contrast the features of the Sal procedure and SAS/ ACCESS software in order to clari fy some common misconceptions about how each interacts with the SAS System and with third~arty database systems, such as DB2, ORACLE, and SOLIDS . In addition, methods of using these procedures together to perform unique and powerful applica tions are presented. Topics discussed include an in depth compari son of PROe ACCESS view descriptors with PROC SOL views, how each procedure interfaces with other databases, and how these two procedures can be used together to create data structures that si multaneously join data from several databases and SAS data sets. The DeLOAD procedure which is often included with SASJ ACCESS software, is highlighted as a means of moving data between databases and the SAS System proc step, create view, dataset, proc sql, create table, select, proc print, sas function, sas library, sas batch, view source 7
88 Dynamic Code Replaces Itself Every 3,000 Executions or Less Kurt R. Luedke,
Jeffrey Keene		 call execute, call symput, proc step, flat file, macro definition, macro variable source 5
89 Econometric Modeling of Sweetgum Stem Biomass Using the IML and SYSLIN Procedures Bernard R. Parresol,
Charles E. Thomas		 A recent development in tree-biomass estimation procedures has been the introduction of the density integral methodology. This technique was developed on the basis of some simple calculus of mass theory. In the initial development of the technique, sectional bole weight was derived from the integral equation formed from a stem-profile function and a specific gravity function fitted independently. Recognizing the contemporaneous correlations between the equations, and the need to incorporate observed stem-weight in the fitting process, a more efficient parameter estimation correlation, estimate, proc step, covariance, proc iml, proc calis, regression analysis source 6
90 Effective Use of Indexes in the SAS System Stephen M. Beatrous,
Karen Armstrong		 Indexes can be used to improve the performance of SAS programs. However, inappropriate use of indexes can actually degrade performance of some SAS programs. This paper discusses how indexes are used and can be abused in Release 6.06 of the SAS system. It also discusses development projects that will improve the use of indexes in future releases and help limit and overcome some of the potential pitfalls of indexes where statement, by group processing, composite key, if-then-else, dataset, subsetting if, proc step, index, cpu time, normal distribution, proc print, put statement, set statement, data step source 10
91 Effective Use of SAS Data Set Indexes in SAS Applications Sally J. Painter By indexing your SAS data sets, you can run certain types of appti cations more efficiently. The ability to index SAS data sets is avail able in Release 6.06 of the SAS System. This paper discusses the costs associated with indexing a SAS data set and the types of applications where the benefits of having a SAS data set indexed outweigh the costs of im~menting and maintain ing the index. The target audience is users who will be designing applications for Release 6.06 and later releases of the SAS System index, dataset, proc step, where statement, cpu time source 5
92 Efficiency Revisited: Large Files and Release 6.06 Juliana M. Ma proc sql, character variable, numeric variable, where statement, dataset source 6
93 Efficiency Techniques for Improving I/O and Processing Time in the DATA Step Neil H. Howard S.avinR m.oney Is uppermost inlhe mindsoC DP man_lien, and the 'SASSystem is certainly. "programmeref!icienl'" too!: miobnat SAS coc::l-e :is .;eqllite4 to perform. trad:lt.iollally code inien:sive tuts. Gratitbtion and -and results are immediate. A familiar quote '51~gge:sb lhal tbe SAS System b euy to Ule. an4 just as easy tD abuse. Becilttl:le ih-:- SAS System is beooming iae:=rea,JngIy popular as II devel(Jpm'!oli.t tool. it IS -8s-tell.tial to evaluate lec:b.a.iQUes fQC i:l;$ eire-clive use. TlUos lu(Dcial will define efficiency, luggeslfa~ots: to be- CQ:Q$ider-ed-in mea:surinB efficiency. and: provide oompariwll$ Q( techniqlle$ Uat imp{UVe (tenciency . A pdmary .tea wber-e the SAS SY$tem <=an be u,ad more effecLively is I/O. It sho-ubl be not(ld that over ~Q" of. the Pr'DCeSSm:g time in any giveD. progralll involves teadio.a and/or vriliDg data (rete-renee 1) proc step, if-then-else, conditional do loop, proc format, put statement, proc append source 6
94 Efficient SAS Software Programming Jean Hardy proc step, cpu time source 4
95 Efficient Use of Numeric and Character Data Types Una Smith A basic understanding of data types in computer pro gramming la.nguages is necessary for the efficient con~ struction of both data sets and programs in the SAS' System. This paper addresses the efficient use of nu meric and character data types at two levels of project development: initial plauning and ongoing mainte- nance. For researchers and planners, some general con cepts of good data design are outlined! with an em phasis on data types; these concepts should help you to design data. sets tha.t are both clearer and easier to work with. SAS users who have not agonized over the choice of numeric or character variables may dis cover that there are good reaso"ns for exercising care in ~his matter. Those who have may find a. review of the major issues helpful as they learn how to use the powerful SAS DATA STEP more efficiently. For the data base manager 1 some pra.ctical strate gies for working around sub~optimal types ate de scribed numeric variable, data step, date variable, sysdate9, put statement, sas programmer, length statement source 8
96 Efficient Use of Resources Using Release 6.06 of SAS Software and SCL Allan M. Kominek Systems evolve. This paper discusses a system that star~ed out as one report and became a collection of 25 programs that were duplicating effort and at times tying up batch queues by contending for the same files. With some help from the author and Release 6.06 of the SAS System, the amount of time spent submitting jobs has been cut by 75% and the amount of CPU usage has been cut by 90%. The new system utilizes SAS/AF'", SAS/FSP, and Screen Control Language proc step, batch job source 4
97 Efficient Use of the SAS/ACCESS Interfaces to INGRES and Sybase Jana Van Wyk Writing efficient SAS applications using the SAS/ACCESS interface to INGRESS or SYBASE is easier If the SAS applications developer is aware of the internal techniques and algorithms used by the database interface. This paper discusses design strategies and performance considerations for users who will develop applications using the SASfACCESS interfaces to INGRES and SYBASE relational database management systems. Included in the discussion of the architecture of the SAS/ACCESS interfaces is the ACCESS procedure, the DBLOAD procedure. the database 110 engine and the Sal, FSEDIT and FSVIEW procedures. This paper is intended for users who will be developing SAS applications using the SAS/ACCESS interfaces to INGRES or SYBASE. This paper is not intended as a tutorial and will therefore assume some working knowledge of the SAS/ACCESS interfaces and the INGRES or SYBASE relational database management systems dataset, proc print, proc step, where statement, order by, relational database, select, format statement, proc sql source 6
98 Enhancements to the GPSS / SAS Compiler Michael A. Greene,
Gretchen K. Jones		 This paper describes Version 3 of the GPSS SAS compiler. GPSS (IBM, 1971) is a popular computer language used for simulating queuing processes. Our implementation of GPSS uses the SAse system to translate and execute the GPSS code. This permits the use of procedures from SAS/STAT and SAS/GRAPH for analysis and dis play of the output data from the simulation. There are a number of enhancements over version 2 (Greene and Jones, 1990) principally including the implementation of a useful subset of GPSS. This version also recognizes SAS statements as GPSS operands so that any SAS function may be used. This represents a sig nificant extension to GPSS, a language which does not contain mathematical or statistical fun ctions. The paper discusses these issues and also how the compiler works symbol table, data step, dataset source 6
99 ERROR: What to Look for When Debugging SAS Programs Brit Minor Have you ever had an experienced SA~ programmer glance over your shoulder for ten seconds and spot an error that you've spent hours trying to find? That programmer probably saw an old friend, an error shelhe's m~de and caught many a time. This paper presents a compendl,-!m of common SAS errors, their warning signs, and suggestions for how to fix or avoid them missing value, sas log, put statement, retain, error, data step, if-then-else, error message, macro quoting function, by group processing source 6
100 Evaluating the Probability Distribution of the Sum of Independent Finite Integer-Valued Random Variables Daniel R. Jeske,
Thomas M. Willis,
Ying C. So		 A fast algorithm for calculating the probability distribution of the sum of independent finite integer-valued random variables is derived. The efficiency of the algorithm with respect to both execution time and the number of required arithmetic operations is discussed. Two examples, including one involving circuit reliability, where the need for this distribution arise are given. The algorithm is encoded in a FORTRAN subroutine that is included in an appendix. It is also shown how, using SASjTOOLKIT""'. a user-written SAS function implementing the algorithm can be created probability distribution, binomial distribution, poisson distribution, sas function, data step, proc freq source 6
101 Evolution and Anatomy of an In-House, PC-Oriented SAS User Group Richard P. Neafus Jr group by, proc freq source 2
102 Exact Confidence Intervals for Estimating Proportions When Sampling from a Finite Population How J. Tsao,
Steven M. LaLonde		 confidence interval, sample, confidence level, binomial distribution, sample size, random sampling, hypergeometric distribution source 5
103 Experiment Design in Engineering Design and Development Douglas C. Montgomery At one time, US industry was the wortd leader in quality and productivity. However, in many markets, this situation no longer exists. American industry has systematically lost its competitive advantage to off shore competition. In particular, the ability to engineer quality into basic product and process design, and then to translate this into a manufacturing system that operates effectively is a characteristic of our competitors that has been partially responsible for this loss of competitiveness. This paper discusses the role of experimental design response variable, randomization source 10
104 Features of SAS/PH-Clinical Software Sandra D. Schlotzhauer Although SAS software has been used extensively in the pharmaceutical industry, our most familiar users have been statisticians and clinical data processors. This paper discusses features of a new SASe product designed to meet the needs of these users as well as other important users, including physicians, CRAs, and managers. The new product, SAS/PH-Clinical software, enables you to quickly create patient groups for special PoPUla tions, query to find out how many patients meet given crHeria, combine across studies, and view summaries for studies. The product provides descriptive statistics Including percentage change and confidence intervals, basic inference, a wide variety of graphs, an ad-hoc tables facilHy, and interfaces to SAS/ASSIS~ software and SAS/INSIGHT'" software clinical data, histogram, clinical trial, pharmaceutical industry, confidence interval, proc means, data science source 10
105 Fitting Sums of Exponentials Using the Method of Residuals and the NLIN Procedure in SAS Software Gail Butler In many applications the user wishes to fit data to a sum of exponentials of the form C = L Aie-ait Initial values for the parameters Ai and Qi can be estimated using the estimate, regression analysis source 6
106 Five Steps to Choosing Optimal Training at Your Site Cathy L. Brinsfield,
Beth Livingston		 sas base, proc step source 5
107 Fun/Aesthetics Jenny D. Wirtschafter conditional do loop, proc step source 4
108 Getting Data from VAX' to JMP-A Way to Handle Sloppy Data David B. Hoffman With the growing trend to pertorm statistical analysis on microcomputers comes the "joys" of moving data from the mainframe to the microcomputer. This paper describes one successful approach used to transfer and format a tourism survey data base from a VAX mainframe to a Macintosh for analysis using SAS JMP"" txt, format statement, dataset, excel source 2
109 Getting Help from SAS/AF Applications Felix G. Meale if-then-else, error message source 3
110 Getting Started with SAS/AF Software Howard Levine proc step, macro variable, catalog entry source 9
111 How Good is Your Data: A Discussion of the Measurement Process Annabeth L. Propst standard deviation, data science source 3
112 How to Enhance Version 5 SAS/FSP Spreadsheet Applications with Version 6 SAS/CALC Software Patricia L. Berryman,
Richard A. Ragland		 Sl-SfCld.C __ Is a new p<oduot In ~e ... 6.07 01111. SAs- System. Converting Veroion 5 SAS/FSF'OI _n'-nl developed _ "'. FSCALC proosduro to a Veroion 5 lonna! allows you to take advantage of tfle changes, "nha.noements, and new features avaK· able In VerSion 6 SASjCAJ...!'j 9Oflvtare. k'llhis pape .. ; we first show an e ..... ple of a oimple Ve",ion 5 PROC FSCALC _1Ioa~on. We "'en take "'at eppIlcation and .how how to do H wi'" SASfCld.C oof\ware. WlII1 tho Version 6 _I!cation, .... show severalonhance menb 10 -the Veraion .5 application that take advantage of some of tho newfea\u",. of Ve,oion 6 of 1110 SI-S ~lem proc step, sas base source 6
113 Implementation in the SAS System of the Bradu and Mundlak Minimum Variance Unbiased Estimator of the Mean of a Lognormal Distribution Sally Powell Like many variables of scientific in quiry, concentrations of pollutants in the environment are frequently assumed to be lognormally distribute.d. Measurements are transformed to the log scale before normal-model statistical regression analysis, lognormal distribution, residual, proc reg, dataset, sample, numeric functions, sample size, proc step, macro variable, confidence interval source 5
114 Improving the FSEDIT Procedure's Data Entry Applications with Release 6.06 of SAS/FSP Software and SCL William E. Dvorak batch processing, proc step source 6
115 Incorporating Adjustment into Sample Data Products Willard Carl Losinger,
Alfredo Navarro		 sample, imputation, correlation, autocorrelation, sample size source 6
116 Inside the SQL Procedure's Query Optimizer Paul Kent This paper describes the methods and strategies that the SOL procedure considers before evaluating queries. Also discussed are enhanc!ltents made to the query optimiser since Release 6.06 of the SAS System source 5
117 Installing the SAS System for Personal Computers Randolph S. Williams This paper should be attended by administrators who wish to use the INSTALL application to produce customized installations of the SAS· System for 0512; Release 6.06 and future personal computer releases. This tutorial steps tlYough the more important features of the INSTALL application, and also discusses other utilities to aid in installations. The INSTALL application is the basis for future installations, inctuding Windows - 3.0 proc step, sample, sas log, txt, sas base, sashelp source 7
118 Intelligent Production Graphic Reporting Applications LeRoy Bessler proc step, macro definition source 11
119 Interaction Contrasts in Repeated Measures Designs John F. Walsh proc glm, repeated measures, proc genmod, proc step, main effect source 4
120 Interactive Terminal I/O Techniques in the DATA Step Using Base SAS Software in the VMS Environment Ken Larsen file statement, filename statement, character variable, sysrc, infile statement, input statement, character constant, macro variable, stop source
121 Introduction to Efficient Programming Techniques Paige Valentine-Query Wilting efficient SAS"" code is a major """""'" of most SAS pro ~HneI s. This paper discusses haw you can improve the effi ciency of your SAS programs simply by cI1ar(jing your progrwnming tedv1iques. In some cases, depencIng on your compuIing environ ment, 0BIt8I1 techniques may -or may not be advantageOUs. After do!emIirWng which programs can be more efficient. you must decide W'hictl methods of programming wiI be most e1fective cpu time, proc step, proc sort, dataset, by group processing, do loop source 5
122 Introduction to Graphic Presentations Bruce L. Green This paper contains introductory suggestions for graphic materials used to support both tutorial and -technical pre sentations. All of us have heard the Chinese proverb, "A picture is worth a thousand words." There is a less familiar proverb, "The tongue can paint what the eye cannot see." tf that seems inconsistent, consider this: the most success ful presentations use visual aids, but sometimes so do the most boring. This paper explains how important vlSUaiS are to a presen tation, and the effects of not using any visuals or using the wrong ones. Also discussed are: text, tabular, bar, 3D and pie charts, as well as color psychology bar chart, data wrangling source 3
123 Introduction to Hypothesis Testing Using the Normal Distribution, the t-Test and the SAS System Arthur L. Carpenter sample, standard deviation, type ii error, null hypothesis, probability distribution, proc ttest, proc step, sample size, normal distribution source 5
124 Is a Picture Really Worth a Thousand Words? James A. Sannerud The abundance of computer hardware and the power of statistical software have simplified the time consuming process of creating presentation quality graphics. SAS/GRAPH provides the SAS software user the ability to produce high quality graphical output. Unfortunately, hardware and software cannot compensate for a lack of skill in graphical design. Unless the analyst has an appropriate understanding of the principles of effective graphical design, the presentation of the data may not convey the intended message. A poorly designed or misleading graph is equivalent to a grammatical error in speech or text. This paper explores the subject of graphical design and discusses both effective and ineffective examples of each missing value, bar chart source 5
125 Job Submission Menu (JSM): A Programming Tool to Standardize Clinical Trial Summarization Iraj Mohebalian,
Mark Becker,
Lora H. Schwab		 A Job Submission Menu (JSM) is described which aids in standardizing the SAS programming that lists, summarizes, and analyzes the data from clinical trials. Design objectives of the system include: self documentation of clinical trial programming, simplified batch and interactive job submission of SAS programs within the IBM/MVS operating system, a library system to document all clinical drug projects, minimization of duplicate programs across clinical studies, potential Computer-Assisted New Drug Application (CANDA) uses. The system was designed wtth flexibiltty to handle both standardized and non standardized programming. After over six years of use the system has gained dramatically in users and is credited with facilitating the evolution of standardization. JSM is compared to one other clinical trial, macro variable, adverse events source 5
126 Life in the SAS Lane: User Experiences Orion V. Hanger proc step, sas log, sas library, sas base, proc cport, dataset, libname statement, proc cimport, proc download source 6
127 LionTamer: Conversion of a Line Segment File to a SAS Map Data Set John E. Dono,
William A. Bosworth,
Jenny D. Wirtschafter		 data step, conditional do loop source 6
128 Load Data Analysis Workstation (LDAW): A SAS/AF Information Systems Application Burtt Blodgett,
Diane Bailey,
John Powers		 dataset, macro variable, sas base, error message, time series analysis, data cleaning source 6
129 Macro Tricks to Astound the Folks on Thursday Morning: Ten Immediately Useful Macro Techniques Bruce M. Tindall,
Susan M. O'Connor		 1l1e SAS" macro language is usefU os a tool for doveIopIng applica tions and lor ensuring ease 01 maintenance. Frequently. however, $AS 8ppIicaIiOI as deovelopers never get theW feel wet"wiUl the macro ~ """"""" !he """" bizsmI fea~ of 1110 language have an ~ reputa1loo. 1l1e purpose ofllU paper 1510 document ten tipjl1Iiat woold be imme<iately .....,fullO 600116 a_ wflen \tiey retun 10 \heir _on lllurOOaymomlng. ThIo paperdooo not discuss \he inII'oOate and oIten 1IIIIcu~ aspocIs 011110 macro laciIIy, such as esoteric macro quoting functions and complex macro variabte 1 efeI encing envi"omIenIs. 1l1e gaol is to offer suggestions for those users who are _ abOut approa<hing the ",sao facilly. 1l1e i11ended aucfience is applications <levelopers with a need for immecIalely appIcolJle ",acro skill. that \tiey COUCI implement when retuming 10 worK after SOOI. Even ""'se who do not know !hey need maoro are imtlted_ INTlIODUCTION Here are ten slmple:speoitIo ways you can make....,... ~ macro variable, macro definition, macro quoting function, sas log, macro facility source 7
130 Macro Windows: Creating a User-Friendly Application Tina Keene Macro wil'lClCJ!t:.ls a!Iow you to create user ~friendly appIcations reqt..ir ing oriy base SAS' software. This paper presents a simple app1ica tion that uses macro windowS 10 inter1ace with ttle uset'o It eontaiAs windows aJIooMng user inlJut !or data set selection. data S<Jbset1ing. and report generation. Techniques !or validation of values are also shown macro variable, proc step source 6
131 Managing Multicollinearity with the SAS System Rudolf J. Freund proc princomp, regression analysis, feature selection, predictor variable, correlation, proc step, proc calis, response variable, forward selection source 9
132 Manipulating SAS Data Sets: An Introduction Richard Alonso This paper and presentation are designed to provide some basic fundamentals on SAS datasets and how they can be manipulated using the SET AND MERGE statements. The intention is to give novice users a guide which can be used as a quick-reference tool. As a first step, it is important to realize that SET and MERGE are only valid with existing SAS datasets. It is necessary to first describe SAS datasets and how they are created before any dataset, proc step, proc sort, by group processing, proc print, set statement, input statement, missing value, merge statement source 6
133 Marie Annotate: How Not to Lose Your Head When Enhancing SAS/GRAPH Output Arthur L. Carpenter proc step, dataset source 5
134 Maximizing SAS Training Joe Dingwerth proc format, proc step, char., user-defined format source 6
135 Micro-To-Host Link Performance Considerations John D. Ebner,
Donald R. Deese		 The SAS micro-to-host link provides the capability to distribute SAS processing between two SAS system environments -- one on a personal computer and one on a host system. For capacity-constrained PC users, the micro-to-host link provides a significant benefit: the ability to manage large SAS datasets on the host, and selectively process small subsets of this data on the PC. This strategy takes full advantage of the host system's capacity and the interactive environment of the PC. However, there are performance considerations when using this distributed processing architecture. First, depending on the communications con figuration, it may be impractical to ship SAS source code, SAS datasets, and other files between the PC and host system. Second, there are several micro-to-host link tuning parameters that affect the performance of uploads, downloads, and micro-to-host system interaction dataset, txt, proc step, distributed processing source 9
136 Mixed Models: A Future Direction Russell D. Wolfinger,
Randall D. Tobias,
John P. Sall		 This paper provides motivation for the use of mixed linear models (i.e. fixed and random effects models) as a generalization of the standard linear model. Several case scenarios from various fields of study are presented illustrating their power and unity. A brief introduction to mixed models theory is given, as well as a review of current software. A new SAS/STAT procedure, MIXED, is introduced and applied to some common examples. KEY WORDS: Mixed Model, Linear Model, Best Linear Unbiased Predictor, Repeated Measures, Co variance Structures, Maximum Likelihood, Restricted Marimum Likelihood, Newton-Raphson proc mixed, covariance, likelihood function, regression analysis source 9
137 Most Creative Use of the Software-Color Perry Watts dataset, proc step source 12
138 Most Creative Use of the Software-Monochrome John F. Schneider proc step, footnote statement source 4
139 MSHAP-The Market Share/Hospital Analysis Program: A Case Study in Application Development with the SAS System Steven C. Gross,
Dennis L. Redfield		 proc step, proc gchart source 6
140 Multiple Engine Architecture: A Comparison of Database Management Systems Using the SAS/ACCESS Interface Alma Zuniga SAS System Version 6 architecture has opened up new ways to interface the SAS System with several database management systems. Engines were a new feature of Version 6 of the SAS System. These engines retrieve data directly from files formatted by other software vendors. This enables you to use SAS procedures and program statements to process data values stored in these files without the cost of converting them into SAS data files. This paper will compare the different database management systems using the SASjACCESS interface proc sql, variable, relational database, dataset, select, proc freq, proc step source 10
141 NeoVisuals, the Movie Anthony R. MacKay,
Douglas Maxwell Masters		 source 1
142 Normalized Data: Implications for the Use of SAS Software, SQL, and Views Patricia L. Gerend Normalization Is the standardly acc8J>ted goal of the database design process because it ensures that data changes. additions. and deletions can be .performed with a minimum of effort and error. To achieve this goa~ however, the normalization process tends to scatter related data into many different tables, resulting in increased effort and potential error for the programmer who must re-connect the data to produce a report. This paper wilt illustrate the concept of data normalization. It win then compare the impact of normalized vs. denorma"zecl data on the complexity of both S~ and Sal code written for data analysis and reporting. FinaRy. solutions will be offered to reconcile the tug-of-war between the needs of data entry and those of data analysis. These solutions- are 1) the creation of standard SOL views to make the data appear denormalized to the analysis programmer and 2) the creation of a set of actual denormaUzed tables from the original normalized ones key, foreign key, data merging, proc step, normalization source 5
143 Novel Use of SAS Macros and the GREPLAY Procedure to Display Multiple Scatter Plots in the VMS Environment Michael A. Walega,
Paula B. Podolnick		 The production of data-dependent graphs can be a time-nsuming and formidable process. Moreover, the display of mukiple parameters, mukiple graphs per page, or a combination of the two can lead to programming nightmares. The graphical presentation of several parameters involves either repetitive code or resubmission of the program. Providing multiple graphs per page necessitates the design of separate TREPLAY statements in PROC GREPLAY for each parameter. To overcome this problem, we have developed a program that employs macros to produce multiple plots per page for any number of parameters without the need to generate repetitive code or hardcoding the TREPLA Y statements in PROC GREPLA Y. Since macros are used to process the data, produce tides and footnotes in PROC GSUDE, axis labels in an ANNOTATE DATA STEP and the appropriate TDEF and TREPLAY statements in PROC GREPLAY, minimal input by the user is required proc step, laboratory data, proc greplay source 6
144 Object Oriented Program Design: A New Wave for SAS Software Robert G. Hall macro definition, proc step, macro variable, %if-%then-%else source 6
145 Old Wine, New Wineskins: Why the SQL Procedure Will Change the Way You Think Peter Kretzman Release 6.06 of the SAS system is NOT Just a "eN release. instead, in the increasingly diverse wood of -software developmenl "sin~ th. SAS Sysl .... Rei ..... 6.00 dramatically ups the ante. by prm'Iding dIff.....t 'and often bell.,.) mechantsms to achieve the same results as some common SAS tools such as the MERGE statement The Sal p",codu",. even 0!'-09 JUGt Q" ordlnruy SAS data sets, Is one of the most e.KdUng new features of Release 6.06. Especially consld .... d In tandem with the new obAIly to Index SAS data sets. PAOe sal immediately jumps Into the category of Mondal",,! Thin9" 10 Know for Ihe advanced SAS programmer. This advanced tutorial compares and contrasts some standard SAS toots (most notabty, MERGE) with equivalent ways to achieve the sam. effects using PRoe SaL Astounding Improvements ... perfonnance can often be achieved by using PROO sal instead of more traditional SAS tools. Concrete examples are provided, along with discussion of efficiency considerations proc step, proc sql, create table, order by, where statement, create view, variable list source 6
146 Optimizing the Performance of the SAS System under OS/2 Carolyn D. Williams,
Julie A. Maddox		 This paper discusses issues involved in optimizing the performance of the SAS" System urder os/~ Specifically. performance options for 0812, SAS software, and hardware are discussed proc step, numeric variable, proc sort, proc genmod source 5
147 PC Performance - How to Upgrade Existing Equipment, What Kind of Machine to Buy Mark Pabst proc genmod, proc step source 7
148 Performance Tuning for Release 6.06 of the SAS System under MVS Amitava Ghosh,
Leigh lhnen		 The focus of this paper is on ways the usef can optimize Release 6.06 of the SAS" System. Emphasis is placed on describing the fol~ lowing concepts: bundling and SAS system options that help optimi zatiOn, memory usage, SAS data libraries under MVS, sorting con siderations, and other methods of optimization options statement, cpu time, elapsed time, dataset, libname statement, sas base, proc sort, memory usage source 5
149 PIAF-A Real End-User Computing Environment Paul Proulx,
Nancy Faille		 This paper describes PJAF (Product Integrity Analysis Facility), a menu driven end-user computing system at Sikorsky Aircraft. a division of United Technologies Corporation. The system provides a fully integrated environment for accessing and analyzing the Quality data that are generated during the manufacture of Sikorsky Aircraft helicopters. The system was developed in 1988 in the end-user community within the Product Integrity department. It has become an integral part of the Quality Engineer's toolkit, providing a common set of analysis tools for multiple company databases. From an initial 2O-user base within Product Integrity. PIAF has grown to support 125 users from 10 different departments. The -system is developed under the IBM MVS system using ISPF as the user interface, TSO CUST for system programming, and the SAS System for all application programming. Extensive use is made of the base SAS software, SAS/GRAPH, and SAS/FSP proc step, date variable, macro definition, numeric variable, bar chart, subsetting if, proc freq, macro variable source 6
150 Picking the Right Informat for Numeric Data Jack Hamilton This poster describes some of the dota formats used by IBM mainframes, and tells how to select the correct SAS Informats to read them. The information in this poster applies to Version 5 and Version 6 of SAS running under the System/370 architecture. IBM Internal Data Formats Six data formats are cammaniy used on IBM mainframes: character, binary, floating paint, zoned decimal, and packed decimal. Here is a brief description of each of them: Name Description Character Character fields are stared as characters - their contents are usually letters, numerals, or punctuation, They can usually be displayed on a terminal or printer. The $111 or $CHARIII informat is usually used to read this data type. Numeric Character A charoder field in which only numeric characters (0-9, decimal point, and minus sign) are stared. This is not really a separate data type, but a subset of the character type. It is the format in which data entry is usually dane decimal, informat statement, character variable source 3
151 Planning for a Smooth Transition to Release 6.06 of the SAS System in Production Support or What Do You Mean, "Not Transparent!" Andrew T. Kuligowski source 3
152 Power Analysis and Instructional Monte Carlo Simulations Richard W. Cole While monte carlo simulations have provided statisticians with a great deal of empirical information about statistical theory, few students have been able to draw upon these resources in understanding a broad variety of statistical concepts. Power analysis is one such example. This paper presents an algorithm written with SAS/IML sortware capable of calculating the non centrality parameter-for a broad class of univariate designs typically analyzed through SAS/STAT software procedures PROC GLM or PROC REG. Its advantages include the capability to handle nonequal cell sizes and ordered effects (continuously distributed interval scale variables). Additionally, this paper shows how one may generate samples of data drawn from user-defined populations using DO-loops and the RAN NOR function power analysis, sample, estimate, macro variable, proc glm, proc step, monte carlo, parameter, degrees of freedom, histogram, sample size, anova, do loop, macro loop source 4
153 PRIME System, A SAS System for PCs Application Used as a Tool to Analyze and Improve Many Aspects of Work Performance Aiman M. Zeid macro variable, dataset, macro facility source 7
154 Problems with Missing Values? Use the SAS System to Generate Replacement Data Daniel R. Bretheim proc freq, proc step, if-then-else source 5
155 Processing VSAM Data Sets Using the SAS System under MVS Amitava Ghosh,
Hugh D. Landis		 TtIe focus of this peper is on the capabiilies of Release 6.06 of the SA5' System in processing ¥SAM data sets. Efr4>/1asis is placed on _bing the fdIowing concept" basic dofinitioos. steps in _Ing with VSAM data sets. SAS apIIoos and ...- for JIIll" 00S6ing VSAM data sets. and SAS operations on VSAM data sets. AIte< soma basic dafinJlIono. _ bo>glno lJy _o.g VSAM external files to the SAS System with the FILENAME statement end wor1<ing with ¥SAM data sets In the SAS DATA s!9p through the INFILE end FILE statemonls, aJoog with SAS automatic VIIIl-.. Examples are given to show reeding. updating, end loading VSAM data sets 1/Yougt1 the SAS System, automatic variable, output delivery system, proc step source 5
156 Producing Publication Quality Business Graphs with SAS/GRAPH Software Andrew Kopras proc step, proc gchart source 7
157 Profiling and Rebundling Medical Care Claims Using the FORMAT Procedure and Macros Michael L. Davis Medical care providers commonly resort to billing separately for procedures which should be billed as a single CPT code. writing a program to profile or rebundle medical claims is complicated by problems such as the: need to evaluate a random number of claims simultaneously permutations of logic used to recombine claims characteristics of profiles or bundles that are not composed of mutually exclusive CPT codes An~elegant solution to these difficulties is to use some of the subtle features of the FORMAT Procedure and Macros. The following describes how to write a program which may be used under Releases 5.18, 6.04, or version 6.06 proc format, audit trail, macro definition source 5
158 Psychometric Applications with Optical Scanning and Analysis W. J. Klein,
Gordon Emslie		 A Test Response System was developed to display and ana· lyse student results on multiple response tests using optically scanned examination forms. TAS combines latest devok>j2ments in optical scanning technology with SAS" 6.04 and SA$" 6.06. A weighting system allows examiners to omit some questions and emphasize others. Files created with optical reatters are saved as SAse files with SAS 6.04 and transmined with SA~ CONNECT for processing with SAse 6.06. The output shows the student's total correct, total wrong, omined, and multiple responses; the student's weighted total; ad justed weighted total according to five levels of guessing penalties; and the studenfs class rank. For each hem the report includes the frequency of each answer (including nonresponse and muhiple responses); item variance; and item-test correlation. Psychometric measures indude Kuder Richardson 20 index of reliability (Kuder and Rtchardson, 1937) and coefficient alpha (Cronbach, 1951) correlation, standard deviation, proc calis, sample, proc freq source 5
159 Regression on Medians of Probability Distributions Martha R. Sumler,
Henry L. Fisher,
Larry L. Hall		 The aedian is a fundaaental parameter in the area of lifeti.e and survival statistics. In toxicodynamics the LDSO, lethal dose tbat results in 50% mortality, is frequently used. The median is also used to describe the incidence of cancer and other disease states. Factors such as nutritional status. age of animal, and exposure to a second chemical can cause the LDSO to shift. It is therefore desirable to determine a functional relationship between the median of a distribution and a cofactor. In this paper we use the SAS. System to examine the use of median regression to predict a continuous dependent variable as a function of a single cofactor and compare these results to the standard ordinary leas~ squares regreSSion techniques. Two data sets were generated using the SAS RANUNI and NORMAL functions. In one, the median line was proportional to the mean line, and both the median and mean had positive slopes with respect to a cofactor regression analysis, probability distribution, proc calis, proc genmod, lognormal distribution, sample source 6
160 Relational Database Design Theory: An Overview Faith Reneé Sloan Wi~ the addition of Strpctured Query Lanquaqe(SQL)1 II. dateb.se 1anquaqe that bas become the American National. standards Institute (AKSl:) standard for rel.atlonal. da~e 1anQUaqe, in V.rsion 6 BAS· software, it 1& ::~;!e~;!f1~~~t Sf: ~r:~~:na\~~f:.b~e a::Si~v~=~~ ~t:t1~~'i1':t~"!.~!;! ~~~~leU: l!oor~!r ~:t;:cJ.~~:!~: the efficient querying Of the sallie by our users. 1:n thia paper 50_ desiqn issues regarding re1ationa.1. database. wi}.l. be discussed relational database, key source 6
161 Remote Access Pulse Recorder (RAPR) SAS and SAS/AF Programs for Electric Utility System Reports Gabor Sebo Monthly and menu selectable RAPR reports are tai lored to the needs of power system engineers and load forecasters. They provide information on critical transformer load cycles, load duration curves, system peak time load distribution, voltage analysis, Kvar flows, power factors. Reports are based on historic load data on fi 1 e and on computed data. Programs compute net generation from gross generation and station service, loading on unmetered transformer wi ndi ngs from metered wi ndi ng loads, utilizing user defined formulae. A user friendly system allows engineers to set up these formulae with relative ease proc means, proc step source 7
162 Repeated Generation of Random Samples to Simulate a Larger Sample Using Arrays and Macro Variables Angie C. Wang sample, macro variable source 3
163 Report Writing Using DATA Step Programming in Base SAS Software David A. Larson,
Charles R. Wartenberg Jr.		 put statement, data step, char., retain source 3
164 Ridge Regression from Correlation Matrix Input Lawrence Hotchkiss,
John R. Gerlach		 This paper describes a sequence of SASprograms that read a fiat file containing a correlation matrix and auxiliary information, then perform a ridge regression from the correlation matrix. The motivation for the project arose from the need of one -of our users at the University of Delaware (Arthur Hoed) to carry out a ridge regression from a published correlation matrix for which the original data no longer existed. All exist ing software at our installation capable of calculating a ridge regression required raw data as input. The paper is organized into three sections. The first summarizes the problem of multicollinearity and the ridge-regression approach to reducing it. The second describes a set of SAS programs which carry out a sequence of ridge regressions, and the last presents some examples of using the programs regression analysis, ridge regression, correlation, sample, predictor variable, proc calis, covariance, standard deviation, response variable, data science source 6
165 Robust Estimator for Linear Regression Ashok Mehta proc step, regression analysis, proc reg, proc transpose, predictor variable, robust estimator, covariance, outlier source 7
166 Running Release 6.07 of the SAS System under UNIX Operating Systems Alissa Schleich,
Renee Ray		 This paper covers aspects of running the SAS- System, Release 6.07, under UNIX· operating systems. It discusses the new X win dows interface as well as converting SAS data sets from Release 6.03 to Release 6.07 dataset, sample source 3
167 Running SAS Software on a PC under Microsoft windows Bruce Nawrocki source 2
168 SAS Applications at the Frankfurt University Hospital Harald Pitz,
Marcus Frenz,
Hans-Peter Howaldt		 proc step, survival analysis, proc freq, regression analysis source 5
169 SAS Dates Simply Stated James C. Stokes How many times have you heard "SAS dates reflect the number of days since January 1, 1960," and acknowledged that fact with a no-problem-l-can-handle-that nod, only to find yourself scratching your head in confusion when you attempt to work with dates in SAS? This paper will track the evolution of a standard date to its SAS form, demonstrate how dates are compared using SAS, and describe some of the common formats used in conjunction with SAS dates sysdate9, input statement source 1
170 SAS Macro Language Features for Application Development Earl R. Westerlund The SAS'" macro language has many lJges. One usa is to facilHate development of applications and systems. This talk· will discuss some features of the SAS macro Ian· guage that are useful to application developers. Topics to be discussed includa: Application development features such as windows, statament style macros, and the autocall facility The haws, whys and whans of macro quoting Options and statements tha! aid in macro debugging. This talk assumes a basic knowledge of SAS macros and macro variables macro facility, macro variable, macro definition, proc print, proc step, dataset, macro quoting function, &variable source 4
171 SAS Macros for Converting Between ARC/INFO Import/Export Files and SAS Data Sets Barbara L. Jackson ARC/INFO is a widely used geographic information system by Environmental Systems Research Institute (ESRQ. ARC/INFO has powerful spatial analysis and presentation capabilities, but its data management and statistical analysis capabilities are limited. The SAS macros described here facilitate exchange of data between ARC/INFO and SAS. The SAS macro, ARCTOSAS, creates a SAS data set from an ARC/INFO sequential export file. The SAS data set will have as many variables as there are INFO items, with identical or abbreviated names and appropriate data types. The SAS macro, aUICKARC, creates a sequential file from a SAS data set. This file can be imported directly into ARC/INFO for use in the geographic information system. All the variables in the SAS data set are included as items in the INFO data set, with the appropriate data types and the same names. The macros run under SAS version 6 for the PC/DOS and VMS operating systems, and version 5.18 or higher for MVS operating systems dataset, variable, call symput, missing value, numeric variable, proc contents, macro definition, proc step, macro variable, %if-%then-%else source 6
172 SAS Report Writing in an HMO Claims Database Environment for Utilization Review Lynn M. Thome,
David Olagunju		 proc sql, proc step, proc format, proc tabulate, dataset, proc freq, sample, insurance claims, date variable, flat file, proc gchart source 6
173 SAS Software and Fractals Kevin Harper,
Karen Falke		 proc step, proc gmap source 6
174 SAS Software and the Complete P &L Approach to MIS Reporting Thomas W. Jack The purpose of this paper is to explain the concept and describe the methodology by which data may be summarized and various typical Management Information Systems (MIS) issues expressed in terms of overall business profitability. This is accomplished via the organization of data into a Profit & Loss (P&L) statement in the format commonly viewed by senior management. This approach was developed over time as MIS manager for large credit card operations, with millions of cardholders. For purposes of example, that application is presented here, although the approach is analogous to all types of business. The following represents a simplified P&L for a credit card operation proc step, credit card, interest rate, proc means, fee income, management reporting source 3
175 SAS Software Automated IC50 Parameter Estimation with Repeated Measures Data Dorene M. Emmrich,
Linda Archer,
Greg Silva,
Charles R. Villarrubia		 proc calis, covariance, regression analysis, proc nlin source 3
176 SAS Software Gets Wet: A Statistical Analysis of New Orleans Weather Joseph E. Earley The purpose of this paper is to present a statistical investigation of climatological trends in New Orleans, Louisiana. Regression and time-series models have been estimated using data on precipitation and temperature for the years 1871 - 1989. Ex-post forecasts are calculated and evaluated for statistical significance and forecast accuracy. PROC REG and PROC ARlMA are the principal SAS procedures used in the investigation. Brief Description of New Orleans Climate The U.S. Department of Commerce National Climatic Data Center describes the climate of New Orleans as follows: 1 The New Orleans metropolitan area is virtually surrounded by water. Lake Pontchartrain, some 610 square miles in area, borders the city on the north and is connected to the Gulf of Mexico through Lake BOIgne on the east. In other directions there are bayous, lakes, and marshy delta land. The proximity of the Gulf of Mexico also has a great influence on the climate proc step, homoscedasticity source 6
177 SAS Software Puzzles '91: The Debugging Continues Warren Repole Jr. dataset, input statement, missing value, set statement, data step, sample, character variable source 7
178 SAS Software with Other Tools on a PC Platform: Experience in a Clinical Pharmacokinetic Laboratory Jerold S. Harmatz Irrespective of ~e platform on which SAS software is executed, personal computers shQuld be appreciaeed as faciiitaeing ~e producti.m aDd r-wming of SAS code~ In the assessment of pharmacokinet1c and pharmacodynamic drug effects, we confront a diversity of statistical and reporting tasks. Seve4al computing tools, that typically are part and pareel of the ~pe~SQnalizatlonn of personal computers, have proven helpful. In this presentation, we review applications of the command shell, of tne editor ~nd of supporting utilities that can be facilely cobbled together in FaBeal and C proc step, macro facility source 8
179 SAS Training on PCs for Students in a Graduate Econometrics Course at the University of Florida Mercedes C. Rosalsky proc step, problem solving source 4
180 SAS Transport Disk Files: Mainframes, PCs, and Minis Tamara Fiscehll Transferring files between different versions of SAS is becoming more important as institutions adopt multiple hardware platforms, or service other organizations with various hardware and software configurations. A SAS transport file is an ideal way to cross software/hardware and operating system boundaries. Transport files can be compressed, thus enabling the "sender" to use diskettes as a more transportable med.ium than tape. Finally, the exis tence of any interactive SAS facility no longer need be a factor in transferring SAS datasets. The process of creating, transferring, and recreating the SAS library is not always obvious, especially when trans ferring files between different SAS versions. Not only does one need to know when and how to do it, but what to keep in mind throughout the entire process proc step, proc copy, dataset, length statement source 6
181 SAS/ACCESS Interface for IMS: Now You See It, Now You Don't Nancy J. Wills,
Louise M. McCrorie		 where statement, dataset, proc step, set statement, format statement, proc sql source 9
182 SAS/ETS Forecasting Application GuI Ege proc means, proc esm source 10
183 SAS/ETS Software in Applied Econometrics: Evaluating Input Separability in a Model of the U.S. Manufacturing Sector John J. Hisnanick degrees of freedom, alternative hypothesis, null hypothesis, significance level, estimate source 6
184 SAS/GRAPH Software - All the Maps You Will Ever Need Jay Jacob Wind,
Brand Niemann		 proc step, proc gmap source 6
185 SAS/OR Software: Solving Real World Problems Keith Weiss SASJOR@ software brings together several solution techniques used in Operations Research. This paper describes several im plemented applications at ICI Americas Inc. which use SASJOR software. We will discuss why we chose to use SASJOR software and what the unique advantages SASJOR software has over other products. Successful implementations of OR techniques can in· crease productivity, improve manufacturing efficiencies, reduce costs and improve profits proc step, data step source 2
186 Scheduling with the CPM Procedure Radhika V. Kulkarni Scheduling with the CPM Procedure Radhika Kulkarni, Operations Research R&D, SAS Institute Inc., Cary, North Carolina This paper gives an overview ofthe scheduling features in SAS/OR@ software. A brief outline of the CPM procedure is followed by several examples illustrating various aspects of the CPM, GANTT and NETDRAW procedures. These examples incorporate features that ate new in Release 6.06 and Release 6.07 (family 3); new options/features in Release 6.06 are illustrated by an asterisk (*) and those in Release 6.07 are illustrated by two asterisks (**). The ability to specify substitutable resources is used to model a course scheduling problem using PROC CPM source 15
187 Screen Control Language - An Invaluable Tool in Applications Development David S. Septoff sas base, dataset source 4
188 Screen Control Language Programming Techniques for Concurrently Accessed Data Mike Carney Whenwrttlng anSCL program that reads Of updatesSAS data sets concurrently wHh olhe, SCL programs or usars, there an> certain lachnlques you may employ when design Ing a program to maximize concurrency Md enSure data Integrny_ This paper explains lhe application of concurrent SCl programming, develops a methodclogy for producing such a program, and Illustrates these tachnlques wHh ex amplee dataset, likelihood function source 9
189 Screen Control Language Versus Macros in Batch Environments Andrew A. Norton macro facility, proc step, macro variable, macro definition, continue, &variable, macro call source 6
190 Selecting Vendor-Produced Training Jean Ussery Many SAse software installations are faced with a growing demand for training. You have two options to consider in meeting this demand: develop your own training materials or select training from a vendor. Developing your own training is time-consuming and costly. Purchasing or licensing off-the-shelf training from a vendor saves time and money. Many vendors offer training on a variety of topics. You'll also find courses in different media, such as instructor led, video, and computer-based. With so many choices, how can you compare the offerings of different vendors and select the product that best meets your needs? This paper outlines a method you can use for comparing different products (even those from different vendors) and suggests some criteria-to aid your selection process. What about developing your own training? If you develop the training, you may be able to produce exactly the training needed by your company sample, dataset source 5
191 SIGNIF and GENFMT: SAS Macros for Computing and Reporting Significant Digits B. M. Horwedel,
J. E. Horwedel		 Reporting generalized statistics such as mean or standard error with excessive significant digits can be misleading and imply more precision than that fOlind in the original measurements. Measurement data seldom warrant more than two or three significant figures to represent values. An appropriate number of digHs is determined by the precision of the reported value. This poster presents two SAS macros that are useful in reporting values with a IimHed number of signfficant digHs. SIGN IF is a macro that will return a value rounded to a specffied number of digHs. GENFMT is a macro that will generate a picture format to be used in printing numbers. Examples using SIGNIF and GENFMT are shown proc format, macro definition, proc step source 4
192 Simple Construction and Modification of Computer Generated Tables Michael A. Litzsinger,
Steven A. Wright		 put statement, dataset, macro call, proc step source 6
193 Simple Regression Analysis and Detection of Influential Observations in Censored Data Using SAS Software Noor Akma Ibrahim,
Isa Daud		 This paper proposes a method based on least squares estimate to estimate the parameters for regressors in censored data. A case deletion method by deleting observation singly and a distance measure were used to detect influential observation. In the first part of SAS program Proe Reg was employed to get the initial values of regression parameter, treating censored observations as failures.With these initial least squares fit. the expected failure time for each censored observation is estimated. These estimates are then used to get a revised least squares fit and a new expected failure time is obtained. This procedure is then iterated until convergence. To make iteration possible within data and PROC STEP SAS Macro was utilized. In the second part of the program after convergence has been achieved each observation is being deleted in turn and the procedure in the first part is repeated until convergence regression analysis, influential observation, censoring, cook's distance, proc step, macro definition source 4
194 SMP/E Exception SYSMOD Reporting Using SAS Software Paul J. Waldowski Th~s paper describes a program written in the SAS system which performs SMP/E Exceptl.on SYSMOD Reportmg for SMP/E Release 4 users. In particular, the program determmes ~hether any SYSMOD~ !hat have already been installed on a MVS system are now exceptIOn SYSMODs by effICIently processing the SMP/E Consolidated Software Inventory (CSI) datasets with the SAS System interface to VSAM dataset, automatic variable source 6
195 SoCalGes Spot Gas Purchasing: An OR Application Which Generates the Input Matrix Based Entirely on User Specifications Deborah Agata,
Martin J. Collette		 proc step, date variable source 5
196 Software Reliability Measurement: A SAS Application James A. McKenzie Information services is a growing industry in the expanding world economy. Computer technology and software continues to playa critical role in the information services industry. 'The reliance on softwart: demands that its validation. verificatio\ and reliability. This paper will demonstrate the uses of the SAS System in the engineering and management of software development projects. Its focus is the use of the SAS System to create tracking databases.. and generate software Rlliability measures in graphical and sample, data cleaning source 3
197 Software Validation and Quality Assurance in the Delivery of Software Systems: A Panel Discussion Judith H. Mopsik,
Liz Poje,
Carl F. Meredith,
Bruce M. Tindall		 quality assurance, proc step source 4
198 Some Motivation and Comments on Statistical Supercomputing Matthew Schall,
Weichang Li		 restricted maximum likelihood, anova, proc step, proc glm, proc genmod source 3
199 SPAM: its Not Just for Sandwiches Anymore Standish Sibley Except for some rudimentary operations. tape processing on IBM's eMS operating system can be frustrating. exacting. resource-inten siva and even potentially disastrous in terms of lost or overwritten files. One portion of Pacific BeU's Survey Project Access Method (SPAM), clscussed in detail here, alleviates many eMS tape-pro cessing difficulties in ways that users of OSIMVS take for granted dataset, proc step, flat file, raw data source 6
200 Statistical Graphics for Multivariate Data Michael Friendly This paper presents an overview of graphical methods for displaying multivariate data. These methods are described in detail in my book. The SAS System for Statistical Graphics. to appear this spring. In particular. I illustrate the design and implementation of custom graphic displays for: adding more variables to a scatterplot (glyph plots) plotting all pairs of variables (scatterplot matrix) detecting clusters (star plots) plotting observations and variables together (biptot) assessing multivariate normality detecting outliers Most of these methods are implemented as general SAS macros, which are included in the book outlier, proc step, macro definition, normal distribution, proc princomp, covariance, sample source 6
201 Structuring SAS Data Sets for Use with the SQL Procedure Lewis Church Jr. key, relational database source 4
202 Subsetting and Ranking Data Using the DISCRIM Procedure Kevin M. Smith proc step, proc discrim, posterior probability, regression analysis, sample source 4
203 Support for Users Groups: What Is Available from SAS Institute Sally A. Roberson For the past two years, SAS Institute has had a formal program for users group development and support. We actively support groups at the local. regional, and intemationallevel. While the Institute sup ports SAse users groups worldwide, this presentation wiD look at our program for groups in North America group by, by group processing source 3
204 SYSTEM 2000 Data Management Software Release 12 Enhancements and Why SYSTEM 2000 Software Is the Best Place to Maintain Your Production SAS Data David W. Pitts,
Clarence C. Talley Jr.		 A detailed review of the latest release of SYSTEM 2000" software enhancements and new features is shown. Major features include expansion of the database file limits for the number of records and components, dynamic file allocation for MVS, ccmplete CIGS C0m mand level support, and full 31-bit addressing support. These fea· tures are viewed along with the SAS/ACCESS· interface to SYSTEM 2000 Data Management Software and the future plans for the engine that make SYSTEM 2000 software a logical storage methodology to be used for production SAS applications dataset, flat file, batch job, production run source 8
205 System Performance and Capacity Planning Using SAS Software Thomas G. Confrey,
Patrick Tierney		 This paper discusses a system for capturing and analyzing data for large IBM 3090 type. mainframes. The data is retrieved from records produced by IBM's Resource Monitoring Facility (RMF). IBM's DF .. 'lMS Data Collection Facilities (JJCOLLECT), and Landmark's CICS Monitor. SAS programs are used to read, summarize, and condense this information and store it in a database. Once slored. the information can be easily retrieved and displayed in various ways (numerically and graphically) to aid in performance management and capacity planning. Some of the performance indicators discussed are: CPU utilization, TSO response time, transaction rates, paging rales, information rdated to active data storage, and CICS response time. Also discussed are various ways 10 capwre and view CPU utilization for computer complexes that run ill multi-image mode (e.g. PRISM). The complete system is available on a public domain lope capacity planning, regression analysis source 6
206 Taking a Bite Out of SAS Software: Jaw Motion Studies on a PC William C. Murphy,
Howard M. Proskin,
Ross H. Tallents		 A time series of digitalized x-ray images of patients opening and closing their jaws is used to create a SAS dataset for use in studies of abnormalities of the temporomandibular joint Using landmarks on the skull as references. SAS macros are employed on the data 10 compensate for the random motion of the patient during the x-rays so that only the motion of the jaw is recorded. SAS software is further employed to calculate various parameters from the data, such as maximum displacement of the jaw and differences between the opening and closing motions. SAS/GRAP~ is used to produce a pictorial display of these motions, and SAS/STA'" is utilized to explore possible differences between the motion of normal and abnormal jaw joints dataset, data science source 4
207 Techniques for Assessing Differential Item Performance on Achievement Tests Delwyn L. Harnisch Evaluation of item characteristics that differentially affect the performance of subgroups of examinees has been and continuos to be of interest to anyone involved in testing. In this paper a statistical method is used to examine the extent to whidl items are interacting with group membership. A SAS macro was developed which was based on the Mantel-Haenszel (MH) procedure. The MH approach provides a chi-square test with one degree of freedom of the null hypothesis that there is no relationship between group membership and performance on the item after controlling for overall perform ance on the test sample, standard deviation, proc freq, effect size, macro definition source 6
208 Techniques for Improving Coding and Processing Efficiency in the DATA Step Herbert C. Reynolds,
Thomas Jacob		 out=, proc sql, proc transpose, proc step, sample source 4
209 TERNPLOT-SAS Creation of Ternary Plots Leon L. Fedenczuk,
Mark Bercov		 proc princomp, macro definition source 8
210 Testing for Cluster Structure before a Cluster Analysis Brian McArdle cluster analysis, outlier source 5
211 Testing for Lack of Fit in Regression Using SAS Software Julie Wildman-Pepe,
Susan C. Schott		 When fitting a regression model to a set of data there are different ways to evaluate the appropriateness of the model. If the observations are replicated on at least one of the settings of the independent variables a test for lack of fit can be conducted. This paper presents the SAS code for partitioning SSE into the sum of squares due to pure experimental error (SSFURE) and the sum of squares due to lack of fit (SSLOF). The F statistic which compares the mean square for lack of fit to the mean square for pure experimental error is calculated along with the p value for this test predictor variable, proc reg, proc step, regression analysis source 6
212 Testing Separate Correlations for Homogeneity and Estimating Common Correlations: Extending the Information from the CORR Procedure Jose Maria J. Alvir,
Jeffrey A. Lieberman,
Gail Lerner		 correlation, confidence interval, proc corr, variable list source 3
213 Texas Instruments' Workstation Interface for Statistical Analysis and Reporting of Data (WISARD) Jim Wyatt data science, sas log source 4
214 The Advantages of the SAS System under OS/2 Stephen Mandel This paper presents the author's experience in working with OS/2 SAS in an environment that includes SAS for the PC, IBM and VAX environments. The discussion includes how well OS/2 SAS interfaces with the OS/2 PCSA and 3Com· network protocols as well as other 0512 applications software. A Beta copy of SAS 6.06 was installed on a QRiO~ 38Gis running IBM OS/2 1.2. The processor is a 25 Megahertz 386 chip with 8 megabytes 01 RAM memory instan9d. There Is no math coprocessor. AU SAS Institute software products were instalJed on this machine including Base"'. SASlSTA1". SAS/FSP", SASIGRAPH", SAS/ETS", SASIASSIS1". SASlIML"'. SAS/OR". SASIACCESS-. SAS/CONNECT-, SASlAF" and SAS/OC'". Compressed SAS/GRAPH maps were also installed. T ogathar I the software required about 65 megabytes of hard disk space. A benehmark program was created to measure performance of the OS/2 version of SAS proc step, excel source 4
215 The Afterlife of a Production System: Software Library Maintenance in a Large Data Processing Environment Elizabeth F. Henrickson macro definition, proc step source 6
216 The Analysis of Two-Period Cross-Over Design with a Binary Response Variable S. Geethakrishnan This paper discusses a two-period crOSS-Qver design problem encountered while consulting. A brief statistical overview of the design Is given and the test statistics used are discussed. Based on this, a macro program Is written to analyze the data for the two-period cross-Qver study with a binary response variable. This program generates a customized table to be presented for testing the carry-Qver and treatment effect treatment effect, response variable source 4
217 The Balancing Act: An Approach to Analyzing Data Integrity During DATA Step Processing Gary L. Katsanis,
W. Keith Skelton		 record count, macro variable, subsetting if, sas log, if-then-else source 5
218 The Best of All Possible Worlds: Part II (The SQL) Frederick E. Pratter elapsed time, dataset source 5
219 The DATA Step Debugger: Bug Hunting Made Easy! Mark V. Schaffer,
Brenda Kalt		 In Release 6.07 of base SAS software for non-U NIX platforms, a full-screen source level debugger will be supplied for the DATA STEP on an experimental basis. This paper is an introductory discussion of the debugger"s capabilities put statement, infile statement, input statement, macro facility, proc print, proc step source 5
220 The Delights and Plights of a SAS Users Group: One Analyst's Story Erleen Arata Duhon dataset, sample source 3
221 The ETRANK Procedure: Procedures for Analyzing Incomplete Repeated Measures Trials with Treatment Related Withdrawals Richard Entsuah,
Edmond Williams		 PROC ETRANK uses a randomization technique to analyze incomplete repeated measures data where the pattern of withdraw is treatment related. the procedures have been developed as direct application to the new techniques developed by Entsuah (Randomization Procedures for Analyzing Clinical Trial Data With Treatment Related Withdrawals Communications in Statistics Vol 19, No. 10, 1990). PROC ETRANK uses the observed full data or the endpoint data, creates efficient scoring systems which are either categorical, time related ranks or the observed levels to generate an empirical significance level (ESL) with time points descriptive statisticsl Operating System(s): All proc step, randomization, clinical trial source 7
222 The Integration of SAS Software in a Multicentric Tumor Registry Harald Pitz,
Hans-Peter Howaldt,
Marcus Frenz		 proc step, dataset, survival analysis, regression analysis, macro definition source 6
223 The Macro as a Tool Janet E. Stuelpner macro definition, macro call, macro parameter, proc step, macro conditional source 5
224 The Null Hypothesis and Regression-Based Predictions James A. Penny significance level, regression analysis, predictor variable, correlation source 4
225 The Obsolescence of the SAS System Revisited Grant Blank N SUGI10 in 1985, I confended that the SAS System .... s clos6 to obsolescence and that significant changes WSle needed for \elSion 6. Now that ",lSian 6 (i.e. Release 6.06) has bHn 181eased, this paper l'8assesses the thesis that the SAS System may be obsolescent. The context of computing has changed dramatically In the 6 ~ since SUGI 10, and statistics and databases have changed as well. Everything has become mote complex, there are many more aooi6nces of users with diffel8nt needs and diffel8nt interests. The SAS System continues to S81W on8 /arpe mari<et _II (/arpe cotpOraIe mainframe US8IS) but, as other groups of users haw emetged. it has been unabfIJ to repeat its mainframe success. Non-mainframe audiences MKJUId be better served by softwat9 more closely adapted to their environment and many mainframe users VtOuld from benefit from easie,.tcruse and more flexible SAS products. The paper suggests Innovaffw new products that can S81W these users proc step, statistical software, macro facility source 6
226 The Pair Chart Macro: A Simple Graphical Representation of the Two-Sample Problem Michael Riggs,
Janna Harris,
Dana Quade		 Tests for equality of means (e.g., t-tests and F-tests) are the principal ones used for inference about the relative magnitude of the scores in one group vs those in another. However, these tests require that the residuals meet fairly restrictive assumptions; textbooks typically illustrate this by depicting normal distributions whose positions are shifted relative to one another on the horizontal axis. Although rank-based tests for shifts in location can be used to examine the same kinds of hypotheses, it has been argued that they have less intuitive appeal than their normal theory based counterparts and are much harder for nonstatisticians to grasp. Quade (1913) developed a graphic device, which he called the ·pair chart-. for illustrating the two-sample location problem for ranked data. Ve present a pair chart macro from which we derive two commonly used rank-based test statistics sample, if-then-else, nonparametric test, macro call, sample size, proc step source 6
227 The PMENU Procedure and SCL: Forging the Missing Link for Pull-down Menus in Applications Kernon M. Gibes PMENUS provide a <:onsislent user interface that is both familiar and easy to use by everyone conversant with the Macintosh andlor MicroSoft Windows system environments. However. there is no direct link between PMENUs (which submit Display Manager commands and not SCL code) and an SCL program. To garner the power and flexibility of SCL as well as the user interface of PMENUs a straightforward method to forge a link hetween these two is explained and then demonstrated by way of an application currently in use at NutraSweet R&D dataset, proc step source 3
228 The Quality Control Process in Applications Development Areas of MIS Scott Maretick **** The purpose of this paper is to. demonstrate the use of the SAS system in the Quality Control Process in App1ications Development. The evolution of the Quality Process, from the deve10pment of a model that illustrates the key re1ationships in the process r to the display of data was facilitated by using the SAS system. The first st~p was to look at the relationship between the measurement variable and the display variable. That is, the relationship between defects per million opportunities and Sigma (st.andard deviations from the ~ean}. This was accomplished using PROC PLOT and pre-determined data points. Once that relationship was determined, the iterative. process of building and validating the 'best fit' model was carried out. Although the original relationship between the two variable.s was discovered to be curvilinear I the best model turned out to be a segmented linear model with the x - v a ria b 1 e logarithmically transformed proc step, proc reg source 6
229 The Real Work in Converting Version 5 SAS Users to Version 6 in a Mainframe Environment Peter L. Rikard This paper was presented to discuss some real world issues in dealing with making the transition from SAS Version 5 to SAS Version 6. SAS Institute has an excellent course that covers many of the details of making the conversion. I recommend thai the appropriate support per.;onneilaIre the course (I did). Many sites are still faced with convincing everyone of the need and timing of the conversion. DEC SysteDl5 This paper is generally addressed to MVS and VM sites. While some of the issues and planning are pertinent to VMS systems many of the problems noted in the conversion do not exist for VMS users. Many of the extended features available to IBM sites have not been available to VMS SAS users. Thus none of the problems with these extensions are problems for VMS users. Choices In the year since SAS Version 6 has become available for mainframe systems there has been a large amount of proc step, proc format source 4
230 The REPORT Procedure: An Introduction and Overview Ray Pass The purpose of this paper is to present an proc step, proc report, proc print, class variable, proc means source 4
231 The REPORT Procedure: Common Questions and Problems Lynn Hannah Patrick As the REPORT procedure adds a new perspective to the task of report writing. new concepts like windowing, templates, and break ing have been added to make report design more sophisticated and user friendly. Adding a new procedure designed specifically for gen erating reports to a system of software that already includes several report writing procedures brings questions as to which method of generating reports is best, or more appropriate for specific reports. This paper addresses the most commonly asked questions from users who have utilized the test versions of PROC REPORT for their applications. It discusses new concepts that have caused confusion and also emphasizes special features that were added to the proce dure to make it the most powerful and ftexible report writing tool in the SAS· System proc report, proc step, proc tabulate, proc freq, proc print source 5
232 The Role of SAS Software in Building a Distributed End-User Computing Environment Robert S. Lohr cpu time, distributed processing source 2
233 The SAS System - A Full-Featured Utility and Language for Programmers Steven J. First proc step, proc freq, proc rank, data step, sample, flat file source
234 The SAS System and the UNIX Environment: You Gotta Luv' It David Wooldridge,
Claire Castell		 bar chart, proc step, proc report, proc print, sas log source 8
235 The SAS System Supervisor - A Version 6 Update Donald J. Henderson,
Merry G. Rabb,
Jeffrey A. Polzin		 by group processing, merge statement, dataset, set statement, output, update statement, sample source
236 The SAS System, Release 6.06: A User's First Impression Bengt Bengtsson At Gothenburg Universities' Computing Centre (GOteborgs Datacentral) we have had the opportunity to test the new version of the SAS System, version 6.06. Our main test environment has been VtNXA TN eMS but we have also had access to the 6.06 version in other environments. What do we think of all the new features? Are they hard to learn? When one has learned to use them, are they useful? What do we like best? Is there anything we do not like? Was it hard to install? Any reaction from the end users? In this paper we will try to answer these questiOns and others and by to share some of the experience we have collected during the test period. Will also take a look at how things did tum oUt. What has happened at our site since this paper was originally written in the spring of 1990 proc step, where statement source 3
237 The SAS System, Release 6.07, and the X Window System: Design &Implementation as Shaped by Market Issues Thomas C. Edds,
Mark Brown		 source 3
238 The SAS System: Interface to DEC'"'s Common Data Dictionary Ken D. Worsham proc print, proc step, dataset, data dictionary, length statement, proc contents, label statement, decimal, set statement, key source 6
239 The Table Macros: A Pathway for Converting the QPRINT Procedure in Version 6 of SAS Software George E. Demuth label statement, proc step source 2
240 The Techniques to Improve Nonlinear Regression Curve Fit Shi-Tao Yeh proc nlin, regression analysis, proc step source 4
241 The TPSTAT System for the Analysis and Reporting of Toxicology and Pathology Data and Its Implementation under Version 6 of the SAS Software Murray R. Selwyn,
Peter P. Gaccione,
Scott E. Moseley,
Virginia M. Ashe		 data science, proc calis, sample, batch job, logistic distribution source 6
242 The Use of the SAS System in Statistical Training William L. Sanders proc calis, proc mixed, anova, proc step, residual, data science source 4
243 The Utter Simplicity of PROC TABULATE Daniel E. Bruns proc step, proc tabulate, proc freq, sas base, proc means source 7
244 The Version 6 Implementation of SAS/SHARE Software in the CMS Environment Ronald Page The purpose of this paper is to discuss the implementation of SAS/SHARE'" software in the VM/CMS environment. The paper's focus is the access methods employed by SAS/SHARE software for data transferal within the VM/CMSenvironment. The first section of the paper examines the communication model implemented by SAS/SHARE software and studies the relationship between a user application and a remote server. The next section of the paper investigates the internals of the IUCV access method. This access method is currently implemented within the experimental Release 6.06 SAS/SHARE software. The details and limitations of this access method are examined. Next, the APPCNM access method is presented. This access method employs APPCNM services that are a part of the 8M architecture for distributed transaction pro cessing developed by IBM~ Lastly, this paper explores the future of SAS/SHARE software as it relates to VM/CMS libname statement, sas library, dataset source 4
245 Three Year Follow-up of a Transition from a Mainframe to a PC Computing Environment Using SAS Software Ray Palmer,
Clyde Dent		 This paper describes a three year follow-up evaluation on the effectiveness of a PC computing environment using PC SAS- on a local area network. This computing environment was prompted by the encouragement of decentralized computing environments by large universities and companies that have typically used mainframes to meet thier computing needs. This trend continues to be seen across the country, primarily because of (1) cost (2) system overcrowding (3) the desire to gain localized control. The University of Southern California Institute for Health Promotion and Disease Prevention (IPR) was one of the first organizations, with a substantially large amount of data and computational needs to make this change in computing environments proc calis, data governance source 3
246 Tools for Conversational Programming in the SAS System: An Overview Roger Staum,
Paulette W. Staum		 proc step, sas base source 6
247 Toward a Unified Analysis Environment: Two Years Later Charles R. Villarrubia,
Greg Silva,
Bruce Ingraham,
Seth Michelson		 raw data, proc means source 4
248 Traverse: The Macros' Macro Edmond Williams Manl~!dIoo with m.IIIlpIe tables within a SAS data base can become a data processing nightmare. Noverthol ... ouch tasks as consolidation, dataa'llalysis. updating, and valldaIIonarelrequently "'qulred to be petfoImed across tile entire SAS data bese. This paper presents a set of macro rouIlneslRAVERSE whicI1 roll ..... programmers from tile painful "hardoCO(la" ropotlllve programming. It offers a powerfU programming tool which simplifies data manI~tlon macro variable, dataset, work library, proc step, macro definition source 5
249 Treasure Trove in the SUGI Proceedings Margaret Menache With about 2000 papers published in the SUGI Proceedings over the past eight years, it is easy to overlook useful ideas and code. Programmers working in many differenct disciplines grapple with the same problems _ how to do preliminary data checking, how to determine what statistical procedures are ar~ropriate for analysis, how to select tables or graphs for effective presentation of the data. Size and structure of the dataset or database can add another dimension to the problem. Many of the presentations made at SUGI describe particular solutions to such issues. This paper provides an annotated bar chart, proc step, line chart, proc greplay source 5
250 Uses of SAS Software for Implementation of SPC on Continuous Processes Robert V. Baxley Jr. SAS· software is shown to be useful for implementation of a statistical control strategy based on the Exponentially Weighted Moving Average (EWMA) for drifting processes. This strategy keeps the process centered on the aim point by making process adjustments while at the same time it detects sudden large shffts for which troubleshooting is necessary. In the past, Monsanto has typically used a modified Shewhart chart with runs rules to determine process adjustments. This new EWMA strategy is shown to be an improvement because of better tuning capability r robustness to incorrect tuning, and a lessened tendency to cause additional variability with large, inappropriate adjustments. A monitoring system developed with SAS/A~ and SAS/OC software is shown to enhance the implementation of this strategy at Monsanto residual, sample, moving average, time series analysis source 7
251 Using a JCL Repository-Informating a Data Center's Production Function John D. Smith,
Kathleen L. Doyle		 proc step, proc print, proc sql source 6
252 Using Advanced Annotate Functions to Display Wind Distributions Ronald Coleman Sometimes there just Isn't a procedure to draw that special graph you need. So you tucn to Annotate data sets with all those functions. But wait, you stili can't draw that graph. What next? Give up? Draw it by hand? Use the advanced functions for Annotate data sets. Here is a graph that draws itself! Rectangles that aren't limited to vertical and horizontal. Variable scaling 10 handle large distributions. A full legend to explain the graph and stili room for titles and footnotes. This special graph is a wind rose, depicting the frequency that winds occur from each of 16 compass points. This type of graph is used extensively by design engineers when structures are planned. The graph Is created by a macro so any user with raw data can use it with a one-line macro call proc freq, macro call source 2
253 Using Experimental Design to Optimize SAS Performance on the Personal Computer Joe N. Hockman SAS software performance on personal computers has recently been the subject of mcreased interest by users on this platform. Maximizing SAS performance necessarily mvolves manipulating both SAS system options and other hardware/software resource options. This paper describes an approach to optimizing SAS performance through benchmarking on several different hardware platforms within the framework of a statistical experimental design. Systematic manipulation of factors such as the SAS system options, disk cache size, use of a math coprocessor, application of a RAM disk for SAS work files, and other resources permits optimization and prediction of SAS performance for any combination of these factors. Results from this study indicate the assessment and interpretation of factor effects and interactions are facilitated by a systematic investigation of key performance determinants elapsed time, contour plot, main effect, proc step source 8
254 Using Hypertext for Online Information Needs in Data Entry Applications Duane E. Ressler,
Dee Stribling		 sample, sas base source 6
255 Using OS/2 Features with Release 6.06 of the SAS System Jeff McDermott The recent release of the SAS@ System under IBM OSI2@ takes advantage of many exciting features. This paper briefly describes those features and also details InterProcess Communication (IPC) methods available using the SAS System under OS/2 including: unnamed pipes, named pipes and Dynamic Data Exchange (DOE). Examples 'of each show how these methods ,allow your SAS session to share data without an intermediary file between other SAS sessions or other third party software. even across a network! dataset, proc step source 5
256 Using Pull-down Menus with SAS/AF Applications Joann High Release 6.06 of the SAS system provides a new facility for entering commands into the SAS/AF application, the PMENU procedure. In previous versions of SAS, the AF applications commands could be isSued from either the command line or by using function keys. The command line allowed commands to be entered and then executed by pressing the Enter key. Function keys allow the command to be assigned to a key, thus simplifying how commands are issued and minimizing key entry errors. This paper is intended to give the reader one of many possible solutions for designing an AF application system that uses function keys for navigation. After reviewing the SAS methodology presented below, the PMENU solutions should reduce the PMENU learning curve and aid in resolving syntax questions for others wishing to use these functions proc step, dataset source 3
257 Using SAS Macro Language for Efficiency and Flexibility in Displaying Data for Clinical Trials Anne Brockley In clinical research, requests are often received for customized data listings, each containing a subset of available information from the database. In addition, requests are received for all the information in the database about an individual patient, or a "patient profile". The process to obtain the same information with separate coding efforts can be tedious and time-consuming. This paper describes a process to produce customized data listings and patient profiles with two easy-to-use interrelated macros. Although written in context of clinical research, using SAS macro language to increase efficiency and flexibility in displaying data is applicable to other disciplines as well put statement, proc step, output delivery system, proc print, clinical research, macro facility, macro definition source 6
258 Using SAS Macros with SAS/AF and SAS/FSP Software in Developing User-Friendly Applications Jennifer C. Culp proc step, macro variable, macro definition source 8
259 Using SAS Software as a Programming Language to Develop Generalized Systems Barbara M. Ison,
Dania P. Ferguson		 The Natlonol AgricuItund SWiotico Serriee (NJI3S) C>JDducta nwl>' .......,. each,.ar alId has DaOdIbr an aut"",.t ed .,.tom to odit, .....u,.... alId .....".....;., !to __ date. WhiIo """"'YO have 1IW\Y """"""" .ttributeo. each ill _ aad recurring ones chaneB from.1'f!1M tO~. Matb.o:.tpMica) ~ at NASS ""'" ...,. fImWiat..nth BAS 00_ SlId _ SAS -..... .. a]>OW8Jful toollbr date ~ Bo-t "" thio, we decidod to use SAS -..... .. a programiDiDg \ansaagI> for """ ...... Survoy p,..,.,....u,g s,.rem. . This paper w.iII deacribo how to ..... BAS ___ to build a goeneraljzod sptem that placoe the po_ of BAS .. 11: ....... diroctl,y in the UM1"$ handL By using FSEDIT*'.IM the main u&aZ" inte..r..:., and developing BAS progr<IIN'that _te BAS ">do, our ...... can tailor the geo.oraIind system Ibr!hair survey.....& withoutboin&"~_themanopmenl:offlleoSlldother ..,.tom details. I. Introduction GmaraIizod. sptemo -.... programiDiDg blip, ...... tralDlDg nooda, anaIyais, deaign macro variable, sample source 5
260 Using SAS Software to Assess Factors Affecting Employee Attitudes and Turnover in Temporary Offices Mark Holdrege,
John Cunningham		 The analysis of data collected from separating employees is extremely valuable in understanding the determinants of turn over and in designing work environments. This analysis is particularly valuable when the environments studied are created temporarily every ten years, as is the case with the processing offices (PO's) established 10 process the 1990 Decennial Census of Population and Housing. Personnel Division developed an exit questionnaire for employees separating from these offICes to compEete and entered the data into lotus '-2-38 spreadsheets. We then converted the data to SASI8 data sets on the personal computer. We used the TABULATE procedure in base SAse soft ware to produce the initial multiple response reports and used SAS/GRAPH8 to structure and present information for prelimi nary analysis by managers set statement, proc tabulate source 6
261 Using SAS Software to Develop a Clinical Studies Summarization System Mary Anne Nemeth,
Teofil Boata		 SYSTEM DESIGN Based on the requirements described above, SAS 5.18 was chosen as the software package to use for this application. The 950 combination 01 SI>S/_, SI>S/FSP", and SI>S mac:tos would provide the tools to quickly develop an application wi1h tile necessary flexibility. The format and contents of the summary reports was determh1ed by Oinical Research, Biometrics, Oinical Communication., and Clinical Programming. Generic SAS code was written based on _ ~ns. Control tables __ created to _ information about tile specific d_ts and variable. tIlat the generic code would use to produce reports. This Information was organized as a series of SAS datasets, and linked to the program code with macro calls and macro variable names associated with the symbolic fields on the /IF application screens. The control tables datasets are as follows: DBA - Database names and associated format libraries. Each database is assigned a number clinical trial, macro variable source 4
262 Using SAS Software to Report Mainframe Utilization and Performance Daniel E. Lodter This paper presents a brief overview of reporting mainframe capacity and performance. It identifies some difficulties that are often encountered when performing the analyses. It also outlines various ways in which SAS Software can be used to process the available data and suggests ideas for reports and graphs that best present the output bar chart, cpu time, format statement, sample, proc step source 6
263 Using SAS/AF Software, SCL, and the SAS Macro Language to Develop an Interactive Archival System Jerry Musial proc step, data wrangling source 6
264 Using SAS/CONNECT and SAS/ACCESS Software to Access Data in a Distributed Environment Stephanie Creech,
Jana Van Wyk		 SAS/CONNECT software offers SAS(J users connectivity between numerous operating systems and hardware configurations, and SAS/ACCESS. software allows SAS users access to data in numerous database management systems (DBMS) on a variety of operating systems. When these products are used together. you can have effective remote SAS processing of the data in a database management system. You can also easily transfer the data from the DBMS between the SAS sessions on the local and remote machines. This paper is intended for users who will benefit from using SAS/CONNECT software and SAS/ACCESS software to develop applications using one or more database management systems in a distributed environment dataset, proc step, select, proc download, proc upload, work library, where statement, variable, unique key, proc sql source 8
265 Using SAS/GRAPH Mapping Features to Display Meteorological Data for Geographic Interpretation and Analysis Thomas L. Puterbaugh,
Ron L. Lundine,
LeWayne Jansonius		 proc step, proc format source 6
266 Using SAS/GRAPH Software to Create Trees that Model Developmental Biological Phenomena Perry Watts,
Samuel Litwin		 Interpretation of clinical or biological data often requires an underlying mathematical or statistical model. Such models can increase the power of experiments to coruum or refute precisely stated hypotheses, but they often require visual displays to explain their construction. Tree structures are especially appropriate for displaying developmental models. Three different types of trees created with SAS/GRAPH R are presented along with supporting code and intennediate graphics output to illustrate selected models and to show how the trees are actually constructed. Altemative visual displays that the trees supplement or replace are also presented. The flISt tree. a dendrogram, uses PROC CLUSTER to create a data set that ranks distances and locates hierarchical clusters. The dendrogram's construction is illustrated by using a distance matrix of mileages between 10 American cities taken from the SASR SAMPLES library sample, proc step, proc cluster, monte carlo source 6
267 Using SAS/GRAPH Software to Produce High-Resolution Color Images for Publication Craig Sampson,
Woody Middleton		 The goal of the Publications Division at SAS Institute is to produce accurate, high-quality product documentation for a reasonable cost and to have that documentation available for shipment with new releases of software. To- accomplish this goal, the Publications Division keeps as much of the prepress work in-house as possible. In particular, we sought and found a sotution to our need to image SAS/GRAPH" output for our documentation. This paper analyzes the imaging problems and then describes how a team of program mers from the Publications Division and the SAS/GRAPH Computer Graphics Division worked together to modify the SAS/GRAPH PostScripr driver to provide a solution to our color imaging problems proc freq, quality control source 3
268 Using SAS/IML Software in a New Approach to Dealing with Multicollinearity Kevin D. Brunson,
Lawrence C. Marsh		 The purpose of this paper is to demonstrate the use of the SAS/IML software procedure PROe IML in a theoretical econometrics project that involves monte carlo experiments and non·linear search algorithms. We first impose decision rules in a principal components setting to compare the outcome of a pre test estimator suggested by Mundlak (1981) against a more traditional mean square error reduction rule. The decision rules instruct the SASIIML software program to iteratively delete components one at a time until an optimum measure has been satisfied. Next we investigate the feasibility of a non-linear algorithm that iteratively searches for the optimum restriction to impose on a classical least squares estimator that will yield the same results as the variance reduction rule used in the monte carlo experiments proc princomp, mse, regression analysis, monte carlo, response variable, parameter, covariance source 5
269 Using SAS/IML Software to Forecast Corn and Soybean Yields Thomas R. Birkett The United States Oepartment of Agriculture uses SAS/IML· software to forecast corn and soybean yields. The statistical model used is an example of a linear model with parameters estimated under a linear restriction. It is also an example of combining several linear models into one global model so that the individual models satisfy a global restriction. The matrix representation of this model is programmed directly into SASIIML software predictor variable, sample source 4
270 Using the ALL-IN-1 Integration of the SAS System under VMS Linda Helwig All-tN-PM Is a Digital Equipment Corporation package that aHows users to access several office applications through the use of menus. Since Release 5.18 of the SAS System. SAS Institute has provided integrations with a mechanism to access the 8AS System without ever having to leave the ALL-IN-1 environment source 11
271 Using the ANOVA Procedure with Summary Statistics as Input David A. Larson,
Charles R. Wartenberg Jr.		 proc step, anova, proc anova, proc glm, proc genmod source 3
272 Using the C Language with SAS/TOOLKIT Software Oliver Bradley This paper is an introduction to using the C language and SAS(fOOLKlT'" software to write your own SAS procedures. The paper uses a popular 1991 SASware Ballot item, replacement of missing values (item 12) to illustrate some important SAS{fOOLKIT concepl8 and functions; The paper assumes that you have some experience with C language and with using SAS procedures. Previous experience with writing SAS procedures is not needed. The example procedure reads a SAS data set and replaces missing values for all numeric variables or user-selected numeric variables. The syntax is as follows: p~oc NOM[SS DATA';npurdaraser OUT'outputdat ** et .NEWAL~=numeI"; VAR veri ablu; !EXCLUDE variables; The defaults will be discussed in detail below, but they are what most users would expect. The procedure is portable to any hardware supported by SAS{fOOLKIT software, with no source code changes required Complete listings of the procedure and the grammar are at the end of the paper missing value, proc step, numeric variable, out=, by group processing, input statement, in=, variable, variable list source 11
273 Using the Concurrency Engine Effectively under VMS Darylene Hecht dataset, cpu time, proc copy, set statement source 5
274 Using the DATA Step Graphics Interface in Screen Control Language (SCL) Jade Walker proc step, bar chart source 4
275 Using the IML Procedure for Iteratively Reweighted Multiple Linear Regression with a Block-diagonal Covariance Matrix Donald H. MacQueen We have been using SASIIML software to model the magnibJde of a seismic event using signals from one to four seismic stations. Geological properties of the earth through which the shock waves travel introduce a correlation between the recorded signals for each event. Thus up to four correlated dependent variables are recorded for e8ch value of the independent variable. We estimate the dependence of the measured seismic signals on the source using weighted linear regression with a block-diagonal covariance matrix in which the blocks are not all the same dimension and the covariances need to be estimated. PROC IML is a useful and succinct programming tool for this problem. The basic definition of the magnitude of an earthquake is m-log(A)+"o where A is the amplitude of the signal recorded by the seismograph and 40 is a correction for the distance between the earthquake and the seismograph. This definition has also been applied to man-made seismic events (for example covariance, regression analysis, response variable, proc reg, proc step, correlation, likelihood function, proc iml source 5
276 Using the PLOT Procedure to Avoid Fear of Curves in Economics Robert Singleton Many students of basic economic principles courses identify as their earliest stumbling block (which many never overcome) a lack of familiarity with graphs and their interpretation. This writer finds that the PLOT procedure can be used as an heuristic device to reinforce the learning of crucial principles in economic (and computer) literacy. Since most economics texts proceed the same way through progressively difficult concepts, the approach shown here, which requires that students demonstrate mastery of graphical analysis, and of SAS Code, to produce plots of several key concepts in economics, can be used in most beginning courses. The PLOT procedure is remarkably easy to understand and use, and has helped many students learn graphical analysis sooner proc step, proc print source 8
277 Using the Power of SAS Software to Manage a Graphical Database Rich Priem,
Anita Clark		 To explore for and produce petroleum reserves. oil companies must lease mineral rights from the land owners. Historically. drafters have maintained complex linens with hand-posted polygonal lease outlines and textual attributes. Reproductions of these maps are then used by land men. geologists. and geophysicists to conduct their respective business. To conserve manpower. and to enhance flexibility and tum-around. it has long been desired to automate this process. As a leader in the industrY. Exxon has invested heavily in the development and utilization of LOMS. the Land Ownership Management System; an ISPF dialog running on an MVS mainframe which utilizes the tremendous power of SAS to input. store. manipulate. and map the multitude of information pertaining to more than 65.000 domestic leases. This paper will present an overview of the "tool box· of SAS routines used to maintain the LMDB (Land Management Database) audit trail, data science source 6
278 Using the REPORT Procedure in a Nonwindowing Environment Helen Wolfson Many users of 1he newest SAS' system repon......rung prooedure, tOo Ra>oRT procedUre. use1he procedure in a windowing environ ment so \hey can modify a "'pori repeate<Iy and see the results immediately. However. you may pn;tfer to 1,199 the- PROC::edure in a nonwindowing envIroomont. whore eIoments of tOo REPORT Ian ~age serve the IKJrpo&e of windows and commands in the window· ing envi"oM1enl. This peper describes how to buiId...-a! difl9rent _ of reports usi1g the REPORT language proc step, proc report source 6
279 Using the REPORT Procedure to Generate Monthly Summaries of Invasive Bacterial Diseases Raymond L. Ransom proc report, proc step source 2
280 Using the SAS System in a Distributed DOS to VMS Environment Using PCSA Software Features Tony Fisher More and more I companies are faced wi th the proliferation of personal computer workstations to meet their computing needs. Even though these workstations provide a nice environment to the end user, they also can have problems, specifically, resource limitations in the amount of CPU power, memory, I/O capacity, and disk space. Also, much of the data required by end users often resides in centralized databases. These data are impractical for storage on a personal computer due to the amount of data and the need to share the data among many people. What would make this environment more appealing to both the end user and the systems administrator? The ability to share data between different systems in the company computer network maximizes efficient use of storage. The ability to develop applications that execute using the most appropriate resources of the network allows the application developer to make the most intelligent decisions about how to use the available resources libname statement, proc step source 6
281 Using the SAS System to Customize Life Table Reports for Clinical Trials Margaret Lee,
MaryAnne DePesquo Hope,
Tenya Marie Economou,
Dorothea Collins		 A computerized method to produce either actuarial or Kaplan-Heier life table results was developed· for the West Haven Veterans Affairs Cooperative studies Program. PROe LIFETEST, PROC PRINTTO. the Macro facility and SAS/GRAPH PROe GPLOT were utilized to facilitate comprehensive reporting for monitoring large cooperative clinical studies. The four steps in the computerized proeess are: executing the LIFETEST procedure. creating a sAS SYSTgH database, producing customized survival tables. and generating survival graphs. The powerful Macro facility provides the mechanism to process survival tables over multiple strata and PRoe PRINTTO allows the stacking of successive outputs from SAS procedures on an external file. Beeause of the generalized nature of the approach this procedure can be adapted to many other reports utilizing SAS output life table, proc step, conditional do loop, macro facility, clinical trial, proc gplot source 6
282 Using the SAS System to Solve a Complex Programming Problem Orion V. Hanger bar chart, data step, proc transpose, proc step, substr function source 4
283 Using the SAS/ACCESS Interface with Relational DBMSs: Beyond the Basics Bradley W. Klenz The SASiACCESS interface software to the relational database management systems OB21111 , SQL/OS1Ii'I, ORACLE@, and RdbNMS provides $AS users with an easy "'Yay to access data stored In rolational database rna_ement system (DBMS)_. After mastering the basic tasks such as extracting a specific subset of data from a retatlonill DBMS table or ins.rtlog new rows Into a l'Glatlona! DBMS table. users ma~ be laced with more difficult tasks such as extracting data from a relational DBMS table usiog key vailles from a $AS data set or de$igning ~plications that allow mleractive access to data m reiationaJ DBMS tables. ThiS paper ~escribas wo.ys to acoompllsh some 01 the more advanced tasks tIlal tJBers of SAS/ACCESS interface software might face. This paper is intenc:ied for users who are already famJllar with the basic use of the SASJACCESS Interlac. 10 either DB2, SQUDS, ORACLE. Of Rdt>'IIMs dataset, proc step, proc sql, where statement, update statement source 7
284 Using the SQL Procedure Kirk Paul Lafler The SQL procedure follows most of the guidelines established by the American National Standards Institute. 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 methods and procedures. This hands-on workshop illustrates numerous examples of-how PROC SQL and its many statements can be used. 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 SOL Procedure proc sql, proc step, create table, order by, group by, relational database, select source 4
285 Using the SQL Procedure in Data Management Applications Merry G. Rabb The intrexiuction Of the Sal procedure In Release 6.06 provides the SAS· applications developer with a powerful new data management tool. When determining what coding techniques are to be used in a production application, you may want to consider PRoe Sal as an alternative to traditional SAS DATA and PROC STEP methods. Using examples based on real applications, this paper compares using PRoe Sal with other techniques and presents guidelines for the use of PROC SOL and Structured Ouery Language (SOL 1 views proc step, dataset, proc sql, cpu time, create table, create view, network data source 5
286 Using the VBAR and HBAR Statements and the TEMPLATE Facility to Create Side-by-Side, Horizontal Bar Charts with Shared Vertical Axes Labels Lela M. Brown PRoe GREPLAY's TEMPLATE facility provides the user with the means to create graphic images that are not available as part of any of the standard SAS/GRAPH procedures. This paper demonstrates a proc step, proc greplay, proc gchart source 5
287 Using the Version 6 SAS/ACCESS Interfaces to ADABAS and to CA-DATACOM/DB Barbara A. Barrett The Version 6 SAS/ACCESS" Interfaces to ADABAS· and to CA-DATACOM/DS'" in the MVS environment have many improve ments over the ADBEXT and DOSEXT procedures in VerSion 5. You can directly access any database file or table without making an extra copy. Both read and update operations are supported. The two-table limit for joining files has been removed. This paper describes the new SAS/ACCESS interfaces and compares their functionality to the Version 5 interfaces where statement, dataset, variable source 6
288 Utilizing RMS File Attributes Using Release 6.06 of the SAS System under VMS Stephen Boccardo,
Lynn Forwood		 Until version 6.06, the SAS system was unable to take full advantage of some of the RMS file attributes available in VMS. This paper will discuss the utilization within SAS version 6.06 of these file attributes, along with a performance comparison. Referencing RMS indexed files, by either primary or seconclary key lookups using the KEY and KEYV ALUE file statement options, will be discussed. In addition, shared access to RMS files will be covered cpu time, proc step, elapsed time source 4
289 Variable Format Report Generator Joy Y. S. Jair,
Rex C. T. Wu		 It is usually not practical to let the end users of an application system write a program to get their own reports,using COBOL or even some 4Gl·s. Especial for some ad hoc demands I time is the critical factor. A user friendly, menu-driven, variable format report generator using SAS/AF I SAS "ACRO facilities is introduced. Through this system, the user only has to select items he needed on the screen. Then * report w ill produce autom~tically. An existing fixed assets system is selected as an example and is under normal operation now. The results show great success. Preface It is not uncommon that the ad hoc report demands from the end users have become the major maintenance load of an application programmer. In order to improve this situation, a user friendly, menu-driven report generator is needed for most of the case. Since it is not practical to let the end users write their own program, we found that SAS/AF combined with SAS MACRO can almost solve the prQblem macro variable, macro definition, macro conditional, proc step, proc sort, sample source 7
290 Version 6 of the SAS System under VSE: You Have to See It to Believe It Catherine B. Corbett The SAS" System and the VSE operating system have both evolved, in fact, burst into the 19905. Previous limftations have been removed and leading edge functionality has been added. SAS Insti tute's SAS System and IBM's8 VSE Operating System can both do more than ever before better than ever before. What about the com bination of the best of both leading edge companies? Well, "you have to see it to believe it"! This paper highlights the exciting changes and enhancements pro vided by Version 6 of the SAS System on the VSE operating system. Emphasis is on those enhancements unique to the VSE operating system with mention made of non-VSE unique enhancements. The interrelationship of new VSE functionality (specifically VSE/ESA 1 to the SAS System is detailed wherever applicable sample, dataset source 7
291 What's New in Version 6 for Batch Users Lisa M. Ripperton dataset, proc step, libname statement, proc report, proc sql, where statement source 6
292 Why Be Normal? Constructing SAS Based Tools to Facilitate Database Design Peter Kretzman Now that the $AS System (as of Release 6.06) features indexed data sets, it's more viable to do real database design using SAS software. This paper presents a number of automated techniques, implemented using SAS macros, to aid in the normalization, analysis. and documentation of SAS databases. These tools. combined with classical database design methods such as entity-relationship diagrams. help you take SAS-based applications out of the realm of ad hoc data hodgepodges and into the realm of production database design. This paper is especially oriented toward SAS software users without much relational database design experience. During the presentation of these macro tools foreign key, key, normalization, dataset, proc step, relational database, proc contents, sample source 7
293 Workload Characterization Algorithms for DASD Subsystems H. Pat Artis During the implementation of DFSMS', the user must develop a variety of constructs and policies that broadly describe physical data set characteristics as well as archive and migration policies for their management. 1bis tutorial will present an algorithm for analyring an existing DASD storage subsystem to identify potential data classes to support the development of data classes for converting an existing storage subsystem to DFSMS control. clustering or principal component analysis statistical teclmiques that have been key in prior workload characterization studies.[4] It occurred to the author that a non-numeric approach to principal component analysis could be used to identify characteristics mexistingdatasets that could be used to develop DFSMS data classes. Like the least common denominator proc princomp, proc freq source 5
294 Writing Efficient Screen Control Language Programs Annette T. Harris proc step, call symput source 3
295 Writing SAS/IML Functions to Call Subroutine Library Functions Richard D. Langston This paper introdu-ce5 the oConeept of interfacing SAS language -code to external subroutine libraties such as the Engineering and Scientific Subroutine Li brary (ESSL) from IBM This interfa<ing is accom plished by de"Veloping user-written SASjIML func tions with the SAS/TOOLKIT produd missing value, record count source 6
296 X Command - Your Gateway to Speedy PC Inquiries Peter S. Dillingham key, proc printto source 5

Back to Conf go 1992 go 1990

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.