HOME QUICK SEARCH AUTHORS KEYWORDS HELP BUGS

SUGI 1990 Papers - Index

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

Back to Conf
go 1991
go 1989
# Title
×		 Authors
×		 Abstract
×		 Keywords
×		 Extras Pages
1 &&&&&&&&&&SAS - Resolving This and Other SAS Code Challenges Warren Repole Jr.,
Michael C. Hein,
Veronica Fasullo,
Kita Garrido		 proc step, proc print, macro variable, macro function, &variable, sample, macro definition source 5
2 A Blueprint for Interactive PC-Based Applications Development Aiman M. Zeid macro variable, catalog, dataset, sas output, proc step, macro definition, macro facility, sas base, libname statement, filename statement, options statement source 10
3 A CMS EXEC for the Calculation of Exact P-Values Using the SAS System Noninteractively Gregory E. Gilbert In many statistical investigations accepting or rejecting a statistical hypothesis is dependent upon the p-value attained by a calculated test statistic. Many tables exist for specification of a critical value of a test statistic that is to be compared to the calculated test statistic; if the calculated test statistic exceeds the critical value chosen (often based on a p-value desired and possIbly a number of other parame ters) the null hypothesis is rejected. These tables can also be used to find an upper bound and lower bound for p-values once a test statistic has been calculated. An exact p-value cannot be deter mined! Many SAS procedures report p-values to the fourth or fifth decimal place and although four or five decimals are adequate for judging Mborderline significance" exact p-values for test statistics with extreme valuescannotbe determined. Often aquickand simple manner would be helpful in determining exact p-values degrees of freedom, chi-square test source 4
4 A Cross-Reference Listing Program for SAS Software Macros Paul A. Thompson proc step, regression analysis, macro call, macro variable source 6
5 A Hierarchical Database Engine for Version 6 of the SAS' System Barbara A. Barrett The first Version 6 SAS engine for hierarchical databases was built for SYSTEM 2000" Data Management Software. SYSTEM 2000 software Is a hierarchical data management system for mainframe computer systems ruMing under IBM· MVS and CMS host sys tems. A hierarchical database represents one-to-many relationships among data records at different levels. This paper describes how to use Version 6 of the SAS System to define, load, query, and update a SYSTEM 2000 database where statement, missing value, dataset, proc step, sample, quality control source 6
6 A Macro-driven System for Background Graph Generation Robert W. Phelps The Executive Financial DataBase of Bowman GraySchool of Medicine of Wake Forest University has evolved into a system containing over 7,000 entities, with over 5,000 graphs produced yearly. To aid in the development of the system, SAS! APt, SASIFSpe, SAS GrapbtD and the SAS Macro Languagee were used. A process for generating graphs in the background was added to aid in the work call symput, macro definition source 5
7 A Mainframe SAS/FSP Mailing List System Carolyn D. Williams,
Tamara R. Fischell		 A mailing list database is an important com munication tool for any organization. Often maintain ing this type of database in a mainframe environment is considered too complicated to be done by clerical staff. However, using SAS/FSP the University of North Carolina Highway Safety Research Center (HSRC) has developed a menu-driven system that is relatively easy to use by most support personnel. The system includes the ability to add, delete, and modify records, as well as output mailing labels. Tracking capability is managed through an ''Update Report." Backup master files are generated before new datasets are created, and fifth generation masters are automatically deleted to reduce wasted storage space. Finally, labels are generated in several different ways using PROC FORMS. This paper will outline the design and develop ment of this system, using examples geared to the beginner-intermediate SAS programmer proc step, batch processing source 4
8 A Managerial Approach to Tracking Staff Assignments Using SAS/AF and SAS/FSP Software William F. Simpson Jr.,
Leonard Bruckman		 proc step, days past due source 4
9 A Methodology for Computer Performance Evaluation in the Apollo Domain/OS Environment Robert N. Bonham This paper describes a quantitative method for performing Com puter Performance Evaluation (CPE) on the Apollo· token ring network. This method describes the monitoring, analysis. and tuning of a dynamic, tightly coupled, Local Area Network (LAN) using base SAS~ software. SAS/GRAPH!] software and the SOL procedure. As LANs grow, quantitative analysis is becoming essential to target future workstation acquisitions intelligently and to improve network performance and end user response time sample, traffic data source 13
10 A Multisite Data Collection and Analysis System Based on the SAS System Raymond L. Ransom,
Allen W. Hightower,
Brian D. Plikaytis		 proc step, dataset, sample source 7
11 A Programmers Tool Using SAS/AF Software Sandra D. Iverson input statement, proc step source 6
12 A SAS Macro Based System for the Preparation of Data for Comparative Bioavailability Analysis Kelly K. Antony,
Laurie T. Elstun		 This paper presents an application of the SAS system for the preparation of data in comparative bioavaiIability studies. These studies compare the rate and extent of absorption of a drug between two or more treatments. Examples of comparisons typically made include tablet vs. liquid, fed vs. fasted, and generic vs. trade. Three macros are described and the end result of their execution is a permanent SAS data set consisting of key pharmacokinetic parameters such as area under the concentration-time curve, peak concentration, time to peak concentration and half·life proc step, regression analysis, proc reg, proc sort, macro call, pk parameter, macro variable source 5
13 A SAS Macro for Testing for Compound Symmetry and the Hunyh-Feldt Condition Richard R. Allen In order for a repeated-measures model to be analyzed by using the compound symmetry, covariance, macro definition source 3
14 A SAS Macro for the Box-Cox Transformation: Estimation and Testing Charles B. Hallahan response variable, likelihood function, proc freq, regression analysis, sample, proc calis, maximum likelihood estimation source 6
15 A SAS Software-based System for Tracking Clinical Study Case Record Forms Utilizing Bar Codes Jules Bosch The volume of clinical study case record forms which accumulates during the life cycle of a clinical study can be considerable; the number of pages may vary from a few hundred t~ 10,000 or more. Identifying and tracking these individual pieces of paper with a manual filing system can be time consuming and tedious, particularly when summary-level information such as a progress or status report is needed. With the application of SAS software products, bar code labels, and a bar code reader. a whole new array of benefits may accrue to those responsible for reporting on the location and status of case record forms, including: a dramatic reduction in the time needed to generate accurate periodic progress reports the ability to produce automated transmittals to cover case record forms being returned to the study sites for clarification the ability to optimize schedules for travel to the study sites for case record forms retrieval clinical trial, dataset source 6
16 A SAS/AF Procedure Using Regression Analysis to Estimate Remaining Recoverable Petroleum Reserves Karen L. Bice,
Kim Wallis,
Bernard Couderc		 This paper presents a SAS/AF@program catalog designed to allow engineers to estimate remaining recoverable oil and gas reserves for a developed petroleum reservoir. Production decline curve analysis and three decline curve models are introduced. SAS/ GRAPH@, SAS/AP, and the SAS@ procedure PROC REG as applied to decline analysis are de scribed. The program presented allows engi neers with no knowledge of SAS programming to manipulate the regression procedures and input data to quickly and accurately examine several valid models for petroleum production decline analysis, generate production forecasts and estimate remaining recoverable petroleum reserves regression analysis, dataset, proc reg, proc step source 6
17 A Simple Method of Producing a Ratio Table Diane Muffett A ratio table can be used to illustrate statistics for various studies including those done on health survey and/or research data. One study might require a prevalence rate; another might need a ratio. In order to generate the numbers in table format, a computation which is essentially a cell by cell division must be done. I found the resources on this topic to be nonexistent. The purpose of this paper is to provide a resource for other programmers encountering this problem. While it will not provide answers for all the tables one might have to do, I hope that it will give the user a basic idea of how to proc tabulate, proc step source 4
18 A Software Interface between the SAS' System and ORACLE' Weilie Qian,
Yuejin Chen		 dataset, proc sql, variable, proc step, structured data, format statement source 5
19 A Solution to Unique Match-Merge Situations Terry Lessmann macro variable, data step source 6
20 A Study in Repetition Resulting in a Utility Macro for Substitution H. Ian Whitlock macro variable, macro quoting function, macro function, macro call, array, macro definition, macro loop, %if-%then-%else source 7
21 A Survival Kit for Academic Consultants or How to Keep Users from Stealing Your Job Sandra Robinson St. George,
Dusty Teaf		 sas consultant, excel source 5
22 A Tutorial on SAS/IML Graphics Linda Binkley SAS/IML" software, an interactive matrix language, contains both a powerful set of graphics primitives and the ability to create cus tomized graphics routines using these primitives as building blocks. This paper explains how to .. use SAS/IML graphics commands for drawing primitives such as lines. circles, points, and polygons .. use SAS/IML graphics commands for placing text on the graphics output .. assign attributes such as colors, line styles, fonts, and fill patterns to graphics output .. generate advanced graphs interactively and incrementally using the primitives * arrange one or more graphs on the display device using viewports .. develop a toolkit of customized graphics routines using SAS/IML modules .. save and manage SAS/IML graphics segments in a SAS catalog bar chart, catalog, vertical bar chart source 7
23 A Tutorial on the NETFLOW Procedure in SAS/OR Software Trevor D. Kearney Trevor D. Kearney SAS Institute Inc. Cary, North Carolina Many problems encountered in industry can be formulated and solved as constrained network models. PROC NETFLOW in SAS/OR software (versions 6.03, 6.04 and 6.06) finds optimal solutions to such models. This tutorial will cover the following topics: network modeling techniques, the performance ofPROC NETFLOW, * new, undocumented or changed options in versions 6.04 and 6.06 source 12
24 Adult Education Student Tracking System Margaret K. Schrempf unique key, proc step source 2
25 Aids in Data Checking William G. LeBlanc Data validation is all too often the forgotten step in the process of data analysis. To spend so much time researching the literature, planning a study, collecting data, and analyzing data without extensively checking the data invites disaster. This paper presents some ideas on how to make data management, specifically the step of validating data accuracy, a little less painful. Included are simple diagnostic techniques, and a description of two programs that can assist in the process of checking and cleaning data dataset, proc step, missing value, proc freq, input statement source 5
26 Alternative Methods of Estimating Piecewise Linear and Higher Order Regression Models Using SAS Software Lawrence C. Marsh,
Manjula Maudgal,
Jaishankar Raman		 The purpose of this paper is to demonstrate how to use SAS0 to estimate a piecewise linear or higher order spline regression model. The method of estimation depends upon whether the join point -is known in advance or has to be estimated and whether a linear or a higher order model is used. When the location of the join point is not known but some reasonable initial guess can be made. the Gauss Newton method may be used to search for the join point as suggested by Gallant and Fuller. For quadratic or higher order models, the Newton-Raphson method provides a useful alternative. These methods are easy to apply to piecewise regression models using SAS@ as demonstrated in this paper. 2. Case where the location of the joint point, X, is known in advance. First consider the problem of a two-segment piecewise linear regression model when the point where the two lines are to cross, X, is known in advance regression analysis, capacity utilization, proc step, proc reg, binary variable, proc nlin source 5
27 Alternative Parameterization for Anova Models Using SAS Software Robert M. Hamer degrees of freedom, null hypothesis, proc genmod, anova, proc step, proc reg, proc glm, estimate, regression analysis source 8
28 An Automated, Intelligent Algorithm for GPLOT Procedure and PLOT Procedure Graphics Axis Scaling Marcia S. Murto macro variable, proc step, proc means, proc gplot, macro definition source 5
29 An Exploratory Data Analysis Approach to Qualitative Response Modeling Using SAS/IML and SAS/GRAPH Software Merwyn L. Elliott,
Ross Hightower,
Caleb Chan		 A graphical exploratory data analysis alternative to qualitative response modelling is presented which can either be a precursor to or a replacement for discriminant analysis and logistic regression. In taking an exploratory proc step, exploratory data analysis, discriminant analysis, data science, mahalanobis distance, covariance source 6
30 An In-House SAS Users Group as Utilized by an Electric Utility Corporation Darlene M. Gerber,
Allan M. Kominek,
Richard T. Hamper		 The SAS'" Users Group at Centerior Energy C6rporation. an electric utility headquartered in Independence. Ohio. a suburb of Cleveland. has de veloped from a single location. one person in charge organization to a multi location, steering committee organization with approximately 200 members. This paper will discuss the evolution of the users group and its current organization along with its goals and advanced sas, proc step source 3
31 An Interactive Front End for a Data Editing Procedure Using SAS Windows Lawrence R. Catlett,
Lance W. Cameron		 macro call, sample source 4
32 An Introduction to SAS/GRAPH Software For New or Nervous Users Neil H. Howard proc step, proc freq, proc gchart, bar chart source 8
33 An Introduction to Screen Control Language with the FSEDIT Procedure Annette T. Harris,
Ann E. Carpenter,
Glenn H. Itano		 The FSEDIT procedure is used to create and edit SAS" data sets and prepare customized data entry windows. In Version 6 of SAS/FSP" software, you can write programs using Screen Control Language (SeL) to increase the functionality of the FSEDIT proce dure as a data entry tool data cleaning, error message, return, proc step source 4
34 An Online System for SAS Information Exchange Using SAS/AF,' SAS/FSP and SAS/SHARE Software Julie A. Smith macro variable, macro facility source 4
35 ANALYSIS OF COVARIANCE: REPEATED IfEASURES AND SPLIT-Plur DESIGNS George A. Milliken repeated measures, covariance, ancova, degrees of freedom, anova, response variable, missing value, proc calis, proc step, proc glm, residual, estimate, proc mixed, proc genmod source 10
36 Analysis of Covariance: Repeated Measures and Split-Plot Designs Gilda P. Pareja repeated measures, covariance, compound symmetry, proc calis, correlation, likelihood function, missing value, proc glm, intraclass correlation, maximum likelihood estimation, proc genmod source 5
37 Analyzing a Regression Model with a General Positive Definite Covariance Matrix with the SAS System Gerardo I. Hurtado This article discusses and proposes a procedure for the analysis of the univariate linear regression model with known general positive definite covariance matrix with SAS/STAT software of the SAS System. Estimation of parameters, hypothesis testing, estimation under constraints and collinearity and influence diagnostics are re viewed. An example is given to illustrate the procedure. KEY WORDS: Linear Regression Model, Collinearity Diagnos tics, Influence Diagnostics, Positive Definite Covariance Matrix. 2 The Model Consider the model equation y=Xf3+ E (1) where X is a n x p known matrix of rank p :s: n, f3 a vector ,of unknown parameters, and € a random vector with E( E) :::: 0 and pd (positive definite) variance-covariance matrix V :::: O';H for some H pd and lT~ > o. In this article we assume that H is known and 0'; is an unknown parameter. The common assumption of the regression model is to consider H :;::: I, but in this article we relax this assumption covariance, regression analysis, proc reg, macro variable, condition index, proc step source 7
38 Analyzing and Reporting Fixed Effects in the Real World Christopher S. McFarland fixed effects, standard deviation, anova, proc step, formatted input, proc printto, correlation, proc genmod source 5
39 Analyzing Balanced or Unbalanced Latin Squares and Other Repeated-Measures Designs for Carryover Effects Using the GLM Procedure D. A. Ratkowsky,
J. R. Alldredge,
J. W. Cotton		 A general approach is presented which makes possible the analysis of any repeated-measures Latin square design, balanced or unbalanced, with a view towards obtaining estimates of carryover effects (also referred to as residual treatment effects), as well as direct treatment effects. Specifying carryover as a "classification" variable enables the adjusted sum of squares for a test for direct treatment effects and for carryover effects to be obtained using standard SAS specification statements in PRoe GLM. Differences between the means of the treatment levels and the carryover levels are readily determined using the LSMEANS statement. The approach may also be applied to many other types of repeated-measures designs including the two-treatment, two-period cross over design and ones in which each subject may receive the same treatment or treatments in several periods. The technique is illustrated using four examples treatment effect, proc genmod, proc glm, proc step, least squares mean, class variable, clinical trial source 6
40 Andrews Function Control Chart for Bivariate Quality Control Meng-Koon Chua,
Mark Reiser		 This paper presents a new approach to bivariate qualtty control with the aid of a oomputer. It introduces the Andrews function and the function's characteristics pertaining to a standard control chart. A proof of the function is shown such that no Type I and Type /I errors will occur. Sample data from Sultan (1984) is used for illustration purposes and a few arbitrary observations are chosen to show the sensitivity of the control chart. Finally. the advantages of Andrews function control chart are outlined sample, type ii error, quality control source 5
41 Aspects of LAN-Based Decision Support Systems Eric M. Klusman audit trail, data science source 4
42 Basic Text Handling with the DATA Step John D. Smith There are many useful things you can do with SAS* DATA steps to handle text files such as source code, reports, or small control languages or setup files that you might invent. This paper suggests why reading and writing text with SAS is useful and illustrates basic techniques with examples that range from reading JCL to processing setup files that control batch programs. SAS fea tUres that would make text-handling even easier are suggested dataset, proc step, proc print, title statement, sample, macro variable source 6
43 Best Presentation of Data-Color Peter Stowe macro variable, macro definition source 7
44 Best Presentation of Data-Monochrome Vincent G. Iannacchione sample, sample size, proc step, proc sort source 4
45 Better Forecasting with SAS/ETS Software Bruce R. Benseman This paper describes our experiences using the SASIETS software when forecasting for a large New Zealand primary producer. This company has an an nual turnover of $1300 million. And it faces large sea sonal fluctuations in the availability and quality of its raw material. The company saves thousands of dollars per day when we forecast these fluctuations better. We will comment on the advantages and problems of different forecasting models and SAS/ETS software. Problem Background This paper describes our experiences forecasting for a large New Zealand dairy company. In the 1988/89 sea son, the company had a record turnover of $1300 mil lion, despite a 7% fall in supplied milkfat. This turnover included $900 million from exported dairy manufactur ing. In this season, the company manufactured about 320,000 tonnes of milk powder, cheese, casein, and butter. And each year, it handles over 33% of New Zealand's wholemilk production proc esm, seasonality source 4
46 Buffering The SAS System VSAM Performance Headaches Michael A. Raithel The Virtual Storage Access Method (VSAM) has had a strong and popular career since it was introduced by IBM in 1973. It is the cornerstone of online applications such as IMS and CICS, and is widely used in vendor packages and batch applications. Its flexibil~y and internal data management have made it an access method of choice. Those who choose to use VSAM can significantly improve their programs efficiency by properly managing the VSAM buffers. There is a specific methodology for determining what kind, and how many VSAM buffers to allocate to a program. If used correctly, this buffering methodology will greatly reduce program CPU time and disk I/O's and lead to better job turnaround time. The SAS System allows programmers to load, read, and update VSAM files. SAS programmers can implement the VSAM buffering methodology via the BUFND and BUFNI .keywords on the VSAM file's INFILE statement sas programmer, cpu time, infile statement source 6
47 Build a Production System in SAS Macro Robert P. Beeman The many nuances available to the SAS programmer can make macros seem daunting and haed to master. But in truth, a very straightforward path leads through the lahyrinth of percent signs and amper sands. At its end is a simple program which will sup port the three basic parts shared by all production computer systems - !be In, 1be Cruru:b, and 11)6 Out. Four uncomplicated macro modules tie together these three processes. In addition to Simplicity, this program maximizes the use of SAS subsystems (for instance, FSP), to shorten code and to'ease later maintenance chores. My emphasis will be upon code that the audience member can take away and use directly in hislher particular application. Build a Production System in macro put, macro definition, macro call, %if-%then-%else, macro quoting function, if-then-else source 4
48 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 SAS" 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 aU the required software packages source 5
49 Calculating the Jonckheere-Terpstra Test Statistic Using the SAS System under VMS. . Steven Bailey The Jonckheere- Terpstra test is a non parametric test for ordered alternatives that is not currently available in the SAS system. This paper reviews the use of the Jonckheere- Terpstra test and presents an algorithm that can be used to calculate the Jonckheere-Terpstra test statistic and its asymptotic p-value. The algorithm is written to be Tun in VMSTM but could be easily generalized to run in a macro environment on other operating systems. The advantages of this algorithm over previously presented methods for calculating the .1onrkhccrc- Terpstra statistic using the SAS system are discussed sample, missing value source 4
50 Calling IBM Programs and Products from C Dick Wiersma SAS/C" Compiler products provide a powerful and flexible environ ment for the development of C programs. Extensions to the ANSI C language described in this paper make it easy to develop applications that take full advantage of IBM" products such as GDDM"· and ISPF source 4
51 CAMS: A System Built on Three Platforms: Useful Techniques Clinton S. Rickards CAMS - the CLearing Account Management System - has components in the eMS, MVS and PC environments. This paper will describe the business probLem solved by CAMS, each of its major components and how they interrelate, and discuss some design challenges that were met. It also illustrates some of the techniques and tricks used to build the system and send data across environment boundaries dataset, proc step source 5
52 CARDS: An Inverted List Database Created Using SAS Software Luke L. Diamond point= option, unique key source 6
53 CBT versus Instructor-led Classes Margaret K. Schrempf Since the introduction of Computer BAsed Training (CST) as an alternative to Instructor-led classes, trainers like myself have had to re-evaluate our Adult course curriculum. Has CST been worth the cost? Does the user or student get solid data processing fundamentals? Do CST students perform better or worse than students who take instructor-led courses? Should there be a marriage of the two? \Nhere does CaT really fit in? This paper will explore the oost of vendor supplied CST vs Instructor CST develop ment, CST course curriculum vs Instructor-led curriculum, and the cost of PC based CST. Rnally this paper will discuss the effect of CST and Instructor-led training on our own students, and some of the solutions that we think work. * Historical Background Each Information Center in every company does not perform the same functions capacity planning, data wrangling source 2
54 Changes and Enhancements to Version 6 Procedures David Driggs This tutorial introduces the new prOCedures available In base SAS" software for Version 6. Special emphasis is placed on the new V5TOV6 procedure for converting Version 5 SAS files to Version 6. New features added to base procedures are also discussed proc step, dataset, libname statement, catalog entry, proc print, proc report, proc format, run, sas log, index creation, proc catalog, modify statement, proc contents, proc copy, catalog, sas library, index, proc compare, proc means source 11
55 Classroom Presentation Techniques Jean Ussery The assignment of teaching others is both challenging and exciting. There are a number of subtle factors involved in presenting information that c-an significantly affect your success as an instructor. Obviously, students must pay attention during instruction in order to retain the information presented. The following proven techniques for establishing rapport, observing and interpreting students' reactions, using classroom aids, handling difficult students, dealing with various personalities, and fielding questions give you the tools you need to meet any teaching challenge. Whether you teach in a classroom environment or on a one-to-one basis, the tips offered here help you teach effectively and help your trainees assilmilate your information readily. The art of teaching is not easy, but with practice and self-evaluation, you can develop your communication and presentation skills so that your students not only learn from you, but enjoy the experience and rate you a top instructor input statement, options statement source 10
56 Comparing Two Files Humberto Mendez,
Peter Russell		 macro variable, character variable source 7
57 Comparison of SAS/GRAPH with Other Mainframe Graphics Software LeRoy Bessler proc sql, batch processing source 2
58 Computation of Variances of Functions of Parameter Estimates for Mixed Models in the GLM Procedure Ramon C. Littell,
Stephen B. Linda		 proc calis, proc genmod, fixed effects, phase i, degrees of freedom, proc step, proc glm, residual, confidence interval, proc mixed source 8
59 Confidence Intervals and Curvature Measures in Nonlinear Regression Using the IML and NLIN Procedures in SAS Software Danny L. Kugler,
E. Henry Lee		 Interval estimates for nonlinear parameters using the linear approximation are sensitive to parameter curvature effects. The adequacy of the linear approximation (Wald) interval is determined using the nonlinearity measures of Bates and Watts (1980), and Clarke (1987b), and the profile t plots of Bates and Watts (1988). These curvature measures and profile t plot calculations are . e implemented uSing the SAS/IML and SAS/NUN procedures likelihood function, proc nlin, regression analysis, confidence interval, sample source 6
60 Conquering the Dreaded Macro Error Katie A. Hubbell The Macro facility is a very powerful tool in SASa programming. By the same token, it can cause some powerful and, at times, intimidating errors. As macros become more complex with increased programmer skill, it becomes more and more important to form a system for writing macros that avoids a~ many pi tfal1s as possible. Debugging must become more sy'stematic and orderly to avoid wasting time and computer costs. This paper outlines some simple macro variable, macro loop, %if-%then-%else, macro put, do loop, proc print, error, proc step, macro definition, macro quoting function, sas base, sample, macro call source 7
61 Constructing Integrated Clinical Information Systems for Traditional and Nontraditional Users Martin J. Rosenberg Increasingly, pharmaceutical companies are attempting to enhance productivity by expanding the scope of their clinical information systems to embrace non-traditional users such as: investigators, in-house clinical staff, and FDA reviewers. As these systems become common place, companies are confronted with the prospect of having up to five different computer systems, each with incompatible data structures: a remote data entry system; a database management system for traditional data entry; a clinical data review system for use by in-house clinical staff; a statistical analysis system; and a computer assisted NDA review system (CANDA) dataset, data review, clinical data, pharmaceutical industry, proc step, data science, clinical trial source 6
62 Consulting, Counseling, and Consoling: The Rest of the Story Linda J. Stewart In the never-ending rush to satisfy the information needs for today's corporations, some type of computing power is being placed on the desks of accountants, engineers, economists, managers, and administrative personnel. Although these employees are experts in their fields, many times they are untrained in the use of the powerful software tools available to them. Technical personnel used to be insulated behind closed doors, but now they are finding themselves on the firing line - at the front of the classroom. In order to survive, these new teachers must put away their core dumps and language reference manuals and learn to understand what motivates adutts to learn. Wnh this understanding, they can be successful in making their classroom experience rewarding for both themselves and the student. A major challenge facing corporations today is how to successfully train and support employees in the use of powerful software tools. U.S source 4
63 Converting SAS Data Libraries under MVS from a Version 5 to a Version 6 Format: When and How Lisa M. Ripperton The advantages and disadvantages of converting Version 5 SAS data libraries to Version 6 format are discussed under a variety of operational circumstances_ Tools for converting the different file types are presented with appropriate examples illustrating mechanisms and ease of use. Conversion strategies are discussed with an emphasis on tailoring these strategies to your particular site requirements libname statement, dataset, proc step, select, catalog, where statement, proc copy, options statement, proc sql source 8
64 Converting SAS/DB2 Applications from Version 5 to Version 6 Bradley W. Klenz This paper discusses changes in the SAS'" System interface to DB2~ between Version 5 and Version 6, with an emphasis on converting applications. The SAS/ACCESS 8 interface to DB2 with Version 6 indudes new features, such as the ACCESS and DBLOAD proce dures and the DB2 interface engine to manipulate data in DB2 tables. In Version 5, SAS/DB2- software used procedures such as OB2EXT, OB2UTIL, and OB2LOAO to access OB2. This paper is intended for users who have written applications for SAS/DB2 in Version 5 and are beginning to work with the SAS/ACCESS Inter face to DB2 in Version 6 proc step, libname statement, dataset, variable, proc append, update statement, where statement source 5
65 Converting Version 5 Files to Release 6.06 Files: Using the V5TOV6 Procedure Helen Wolfson This paper presents a simple introduction to the new V5TOV6 pro cedure. It explains when you need to convert SAS" files from Ver sion 5 to Release 6.06, why you might want to convert a file even if you do not need to, and when you should avoid converting. This paper contains examples that illustrate how to perform a variety of conversions, ranging from the conversion of individual files and cat alog entries to the conversion of an entire SAS data library libname statement, proc step, catalog entry, select source 5
66 CPU Load Forecasting Using the ARIMA Procedure Robert Hubata Computer capacity planners sometimes use . regression to correlate CPU resource demand w1th a natural business unit. With such an approach, load forecasts can be made on the basis of business units, which in turn can be derived from business plans. This paper describes an time series analysis, proc arima, regression analysis source 4
67 Creating a Variable for Collapsing a File Using the FORMAT Procedure Marianne L. Whitlock This paper demonstrates a method of adding a collapsing yanable to a data set. The collapsing vanable assigns multiple observations of a data set to one group and allows summarizing the data set by that . vanable. The vanable denotes specific comblnatlon(s) of any number of other variables. The class variable, proc format, proc step, put statement source 4
68 Creating and Managing an SQL'"Database with the SAS' System Mark Stranieri A database consisting of Sal procedure tables (SAS" data sets), PROe SOL views, and SAS indexes under Release 6.06 of the SAS System has been established. The process of creating and manag· ing this Sal ~ database is discussed. Creating tables and views is best accomplished by using a combination of PRoe SOL state ments and aSAS DATA STEP. In addition, creating indexes on exist ing tables and normalization, or reducing data redundancy, are used to increase performance and optimize queries. Table relationships are established by defining primary and foreign keys. An integrated data dictionary is created to promote and enhance database man agement. Management of the database is provided for by queries, including the operations of select, project, and join. Data definition and data manipulation operations such as create, alter, insert, and update must also be performed to define and fine-tune the data base data dictionary, key, proc step, dictionary tables, create table, foreign key, relational database, format statement, sas library, proc sql, normalization, sas output, dataset source 6
69 Creating PMENUs for Application Windows Holly Whittle This paper describes the new PMENU procedure, which defines PMENU faCilities for windows created with SAS· Institute software products. The paper presents a simple introduction to building action bars, pull-down menus, and dialog boxes proc step, catalog entry, proc print, sample source 5
70 Criteria for Moving a SAS Application to SYSTEM 2000 Data Management Software Carl F. Carpenter SYSTEM 2000" Data Management Software is currently available to SAS" users on IBM'" or IBM compatible mainframes only. Many SAS applications can take advantage of SYSTEM 2000 software options to improve processing performance, DASO utilization, and access to their SAS data. With the availability of the SAS/ACCESS" interface to SYSTEM 2000 software in Version 6 of the SAS System, there are virtually no limitations on access to data stored in a SYSTEM 2000 database. This paper specifies criteria that enable you to identify SAS applications as candidates for utilizing SYSTEM 2000 software as well as the steps that must be performed to accomplish the move of the applications to a data structure based on SYSTEM 2000 software dataset, length statement source 6
71 Current and Future Enhancements of Version 6 of the SAS System under CMS John W. De Boskey Version 6 of the SAS System has a fresh, new look. Increased interactivity and a "point-and-click" environment will help novice users as they create their first SAS applications. Expert program mers can also realize an increase In productivity due to the addition of pop-up windows, added language syntax, and Increased proce dural functionality. In addition to the features currently induded in Release 6.06 of the SAS System, there are many interesting proj ects scheduled for future releases of the SAS System under eMS. This paper discusses some of the current enhancements and future projects scheduled within Version 6 of the SAS System under eMS filename statement, sas library, catalog, txt, proc sort, proc step source 4
72 Data Center Service Level Management System David B. Horne,
Raymond R. Messmer		 In the past, a bank's dependability may have been gauged by the size and invulnerability of its vault. Today a large financial institution may just as likely be judged by its customers, especially its corporate customers and affiliate banks, based on the reliability of the bank's data center. The bank's computer handles inquiries, records transactions, and transfers funds. It must be "up" as much as possible to keep the customers satisfied and allow the bank to keep the customers. Pittsburgh National Bank has chosen to use the SAS System to create an application to track all Data Center problems, monitoring "down" time, cause of outage, impact to customers, etc. This menu-driven system uses SAS/AF" and SAS /FSP" for data entry and maintenance, provides availability and service level reports for management, and also incorporates FSLETTER to automatically produce service level agreements for Data Center customers dataset, data wrangling source 6
73 Data Management Procedures for Large and Small Projects Juliana M. Ma dataset, title statement, sample, proc step source 6
74 Data Reduction and Summary Statistics in the DATA and PROCEDURE Steps Efficiency Considerations Michael C. Hein,
Judith H. Mopsik		 proc step, proc means, by group processing, proc print, proc freq, data step, proc tabulate, cpu time source 8
75 DB2 and SAS/DB2 Software EXPLAINed Diane E. Brown The use of relational database technology is growing rapidly. IBM'S relational database, OB2, is widely used in large IBM mainframe installations. systems developed using relational technology are primarily used for data reporting and analysis. These functions are normally performed by high level languages such as the SAS System. SAS/DB2 Software is the interface between the SAS System and DB2. Users should be aware of DB2 performance considerations prior to running a SAS/DB2 program to extract the data required for reporting and analysis. This paper will use SAS/DB2 software to examine DB2 performance issues proc sql, proc step, select, dataset, where statement, key, proc print, foreign key, relational database, variable, delete, flat file, update statement source 6
76 DB2 and the SAS System - Information Delivery for the 1990s Darius S. Baer As IBM's strategic product for relational databases, Database 2 (QB2H,,) excels by handling very large volumes of data, interfacing with on-line programming languages such as CICS, and ensuring data integrity through automatic backup and recovery techniques. No other database product offers the benefits inherent in DB2. ~ is recognized for its superb data management and reporting capabilities that facilitate enduser information delivery. SAS also offers a well designed, easy-to-use interface between DB2 and SAS, SASlDB2, which expedites movement of data from DB2 into desired reporfs with minimal time expense and maximal product delivered. By emphasizing the strengths of DB2 and SAS, suc cessful information delivery can result. The DB2 database must be structured and designed to take ad vantage of the relational aspects of DB2 by using in dexes and enforcing unique row identification. SAS code must be written to operate efficiently in the manipulation and summarization of data relational database, data wrangling source 4
77 DB2'"Performance Management: A Strategy Using DB2PM and SAS Software Don Baker elapsed time, proc sql, cpu time source 7
78 Debugging Your SAS Macro Application under VMS'": A Practical Approach Jodie M. Gilmore,
Linda Helwig		 One of the most exciting features of Release 6.06 of the SAS· Sys tem under VMS* is the full macro facility. As VMS users take full advantage of the macro faclDty. It Is necessary for them to know how to efficiently debug their macro applications. This paper uses a hands-on approach to present debugging techniques and tools that are useful in macro applications running under VMS. The audi ence for this paper is programmers who are familiar with the SAS language and have a basic understanding of the SAS macro lan guage macro variable, filename statement, dataset, macro function, options statement, automatic macro variable, mlogic, symbolgen, mprint, macro definition, error message, macro facility, put statement, macro put, macro quoting function, &variable, autocall macro source 7
79 Determining and Profiling MVS System Activity with MXG and SAS Software Paul J. Waldowski T~i~ pa.per .describes a program written in the SAS system which determines and profiles daily resource ubhz~t:on In terms of .hourly service unit consumption in an MVS environment. The program utilizes Mernll s Expanded GUlde (MXG) software as a frontend along with base SAS software, the SAS macro language, and SASIGRAPH@ software. Subsequently, each MVS installation is provided with a high-level view of actual system activity sas base, cpu time, raw data, datetime source 6
80 Determining Sample Sizes Needed to Achieve the Specified Confidence Levels - Binomial Distribution Approach Shi-Tao Yeh sample, sample size, confidence level, binomial distribution source 4
81 Determining Weight Values for EWMA Control Schemes Using the ARIMA Procedure John C. Brocklebank moving average, proc arima, proc step, proc calis, estimate, first difference, proc esm, correlation, acf plot, residual, time series analysis source 10
82 Developing Release 6.03 SAS/AF Applications as a Statistical Teaching Aid L. M. Harschnitz,
Laurie A. Abell		 sas programming, data science, dataset, sas output, regression analysis source 6
83 Developing User Applications Using the FSVIEW Procedure in Release 6.06 SAS/FSP Software Scott L. Bass This paper discusses the use of the Release 6.06 SAS/FSP FSVIEW procedure to develop end user data entry applications. Some topics to be discussed are: Comparing the FSVIEW and FSEDIT procedures Customizing the FSVIEW display Using formula entries to create enhanced displays Using Screen Control Language (Sel) in formula entries Accessing SAS/AF entries within FSVIEW Using the WHERE processor within FSVIEW Sorting data within FSVIEW Creating new data sets within FSVIEW * Modifying the behavior of FSVIEW by setting the control level proc step, missing value, proc sort source 7
84 Developing User Applications with SAS/AF Software for PCs: The SAFARI Extract Generator Experiences Michelle Fine In today's changing DP environment, greater emphasis is being placed,on providing users with tools allowlng them to perform functions tradition ally handled in a systems department. As a result, there is a need for soft ware products that facilitate the development of these end-user, applications. The screen deslgn capabilities of the SAS/AF software for PCs along with its error handling facilities and dataset manipulation functions make it a strong candidate for systems of this nature. This paper addresses a real-life . business problem that was solved wlth SAS/AF software for PCs. It beg~ns by presenting a background descrlp tion of the business environment and the existing problem. It then moves on to outline the stages of system development highlighting features of SAS/AF that' were helpful .. The II f~nal portion of the paper provldes tlPS and techniques II for using SAS/AF and offers recommendations for future development efforts macro variable, cpu time source 6
85 Disk Inventory and Control System: A REXX Tool for the Management of SAS Databases under CMS Arthur L. Carpenter raw data, proc genmod source 6
86 Displaying Multiple Graphic Output Mark E. Mitch Displaying mu~iple graphs or graphic output on a single page is a powerful and efficient presentation tool but unfortunately is often difficu~ to program. This discussion addresses several aspects of the methodology and problems of image distortion associated wtth programming these types of graphics usin§. SASjGRAPH- software. Techniques presented in SAS documentation and at previous SAS Users Group Intemational (SUGI) conferences have discussed various methods including the use of SAS procedures PROC GREPLAY, PROC GPLOT overlays, and the use of the SASjGRAPH Annotate facility. Creating a page of mu~iple graphic output is especially challenging when using graphics generated from different procedures (GCHART, GPLOT, G3D, etc.). A SASjGRAPH graphic page can be visualized as a two dimensional coordinate system made of X and Y posHions (Le., HPOS and VPOS attributes) proc step, proc greplay source 5
87 Easier User Interface with SET and MERGE Macros Cheryl C. Hammond-Puhi macro variable, set statement source 2
88 Effective Use of SAS/CONNECT Software Cheryl A. Garner SAS/CONNECT affords users connectivity between numerous operating systems and hardware configurations to allow remote SAS processing and data transfer. This paper discusses some of the more effective ways to use the SAS/CONNECT product to help automate and simplify access to remote data source 7
89 Effectively Grappling with Graphics Chris Potter proc step, proc gchart, proc gplot, bar chart, regression analysis source 10
90 Effectively Using the FSEDIT Procedure Howard Levine proc step, dataset, data cleaning source 10
91 Efficient Data Summary: A Comparison of PROC and DATA Step Techniques Sandra D. Iverson,
Susan J. Slaughter		 proc step, proc means, proc tabulate, by group processing source 4
92 Efficient Use of SAS Institute's Technical Support Sally J. Painter In order to effectively support the SAS· System at your site, you must be familiar with the support services provided by SAS Institute. This paper discusses both types of support available: telephone support the Online Customer Support Facility (OCSF). Utilizing these services to their best advantage will make on-site support more efficient. In addition, Institute supplied tools, such as the Usage Notes data set and the sample library, are discussed sample, sas base, macro facility, options statement source 3
93 Enhancements to the User Interface in Version 6 of the SAS System under MVS and VM Phil Herold This paper describes the enhancements to the user interface in Ver sion 6 of the SAS" System that are specific to the MVS and VM operating systems. Definitions of the various types of 3270 text and graphics terminals lead to a discussion of the device drivers used by the SAS System and the characteristics of each driver. Special device support for the IBM'" 3290 and 3179G/3192G terminals with an attached mouse is described. Also presented is a complete list of mainframe-specific SAS system options and global SAS Display Manager commands. Common User Access features in the SAS System are described, and numerous hints are provided for interacting with the SAS full-screen windowing system options statement, rolling window source 4
94 Enhancing SAS/GRAPH Output with the Annotate Facility Steven J. First The SAS/GRAPH Annotate facility allows the user to enhance graphics created with SAS/GRAPH software. This paper presents the basic concepts of the Annotate facility and introduces the audience to text placement, symbol placement, line drawing, polygon construction, map labeling, custom legends, and line types. Also covered are SAS Institute provided annotation macros and advanced functions to allow complex labeling and freehand graphics proc step, proc gplot, bar chart source 10
95 Enhancing Your Graphics Output Using New Features in Version 6 Randy Poindexter Users of SAS/GRAPH@ software in Version 6 of the SAS" System will find that most of their programs developed for Version 5 applica~ lions will run without modification. However, a number of new fea tures have been implemented that expand the capabilities of the software. Along with these new features, Version 5 to Version 6 compatibility issues will be discussed proc step, run, catalog entry source 6
96 Error Bars with Line Graphs and Bar Charts Shuching Shaw,
Bernard Most		 It is useful to have full control of the ploning of error bars when displaying statistical data. For example, a problem often arises in overlay plotting of two or more curves which iJlustrate mean response and confidence intervals over the same set of abscissas("times"): the confidence intervals may overlap leading to graphic clutter and/or ambiguity in inteIpretation. An algorithm which removes such clutter according to well defined mles and enhances interpretability is described and i11ustrated. Implementing the algorithm entailed development of code to plot the data using PROC GPLOT in SAS/GRAPH which leads to other benefits. Disclaimer: Although the research described in this document has been supported by the United States Environmental Protection Agency through contract number 68-02-4450 to NSI Technology Services Inc., and subjected to Agency technical peer review, it does not necessarily reflect the views of the Agency and no official endorsement should be inferred confidence interval, proc step source 6
97 Establishing SAS Software at a Large Corporation William E. White source 7
98 Estimating Standard Errors for a Proposed Survey Samuel M. Slowinski,
Hank Leddon		 regression analysis, proc calis, sample, covariance, macro variable, sample size, proc step, proc reg, macro definition, savings account, proc nlin, estimate, %if-%then-%else source 6
99 Estimation of Area Under Concentration - Time Curves (AUC) Using the LAG Function and the Trapezoidal Rule R. A. Rogers,
W. L. Tippitt		 Using the LAG function of base SAse software to implement the Trapezoidal Rule, concentration-time or 3imilar x-ycoordinate data de5Cfibing a curve can be integrated to detellTline the total area under the curve (AUC) values. A series of trapezoids are created and the area of each is summed to give the approximate total area. This routine may be applied to any two dimensional data set. The simplicity and relatively small margin of error associated with this method make it preferable to other AUC calculation methods which require more information lag function, data science source 1
100 EVAL: An Application of SAS/ASSIST Software to Forecast Evaluation Philip E. Friend,
Linda P. Atkinson		 regression analysis, residual, proc step source 6
101 FILE and PUT Statements David A. Larson put statement, proc print, proc step source 3
102 FocusEZ: The User-Friendly, Full-Screen, Menu-Driven Utility for Using SAS Software with FOCUS Databases Standish Sibley dataset, proc step, variable source 7
103 Formatted Tables: SAS Macro Programs Will A. Sullivan,
Sharon M. Haney,
Harvie Chang,
Gary Stephens		 proc step, proc glm, response variable, proc genmod, macro variable, proc sort, proc freq, macro definition source 6
104 Fun and Games with the LENGTH and LABEL Statements Standish Sibley proc step, label statement, proc print, proc contents source 4
105 Future Directions for NeoVisuals Software Rick Edwards proc step, batch job source 7
106 Getting the Most from Your Release 6.06 SAS Session under OS/2 Jeff McDermott proc step, dataset, sas base, filename statement, data step, error message, proc cport, sas log source 5
107 Gracefully Handling Empty Files in the SAS System Matt Kleinosky,
Orion V. Hanger		 macro variable, dataset, _eof_, error message, record count, set statement, abort, macro definition, %if-%then-%else, proc step source 4
108 How SAS/AF Software, Other Version 6 Products, and Screen Control Language (SCL) Can Help You Make a Presentation (and Test Your Country Music Knowledge) Julie A. Smith bar chart, catalog entry source 5
109 How to Get the Most Out of Your SYSTEM 2000 Data Management Software David W. Pitts,
Roy D. Vance		 The Version 6 architecture has opened up new ways to interface the SAS'" System to SYSTEM 2000~ software databases. This paper shows you how to take advantage of some of the powerful features of SYSTEM 2000 software. Such features include Coordi nated Recovery, audit trail update log, Multl-User~ software, higher level language support through PLEX, security at the individual com ponent level, and SYSTEM 2000's own native Self-Contained Facility (SCF) languages. SYSTEM 2000 software can be used in conjuction with Version 6 interfaces or as a stand alone data management software system that provides an easy and convenient means of rapid update and retrieval of data sample, dataset source 5
110 How to Stop a Simulation Michael A. Greene,
Gretchen K. Jones		 One reason for simulating a stochastic process is to obtain an estimate for some parameter of interest. In the queuing context for example, the parameter may be the average waiting time of a customer in a queue. This statistic would be estimated from observations of customer waiting times in a simulation. Obtaining valid interval estimates, however, is not straightforward because observations are not independent. Consequently, it is difficult to determine in advance how many observations to make so that a confidence interval of a desired width may be obtained. This paper describes the implementation of one procedure for obtaining valid interval estimates for a parameter. Once such estimates have been obtained, the simUlation may then be terminated. The method used is the regenerative confidence interval, data step, correlation, sample source 6
111 Hypotheses and Sample Size for Showing Equivalence James J. Ashton If you are conducting an equivalence trial, there Is an approach within the Neyman--Pearson theory of hypothesis testing that refor muates the hypotheses. with equivalence of treatments as the after· native rather than the null hypothesis. This approach swaps the roles of the type I and type II errors. It also means you can explicitly control the probability of making the more serious error of finding no difference In treatments when in fact the standard Is superior. Most clinical trials comparing two treatments are conducted to deteonlne H one treatment Is significantly different hom the other. The traditional approach to this problem tests the null hypothesis that the success rates for the two treatments are equal against a two-<lided afternative that they are not equal sample, sample size, type ii error, null hypothesis, type i error source 5
112 If an Axis Is Too Long for the PAGESIZE=, LINESIZE=, HSIZE=, or VSIZE= Options PANELIZE Routines Warren Repole Jr. proc step, lookup, bar chart, by group processing, proc gchart source 6
113 Importing Data from External Files to NeoVisuals Software James Richard Hagen The application of three-dimensional graphics has evolved into two distinct categories: images based on ideas, concepts, and imagination; and images based on data from tangible, measurable objects. This paper addresses the processes involved in the second area; importing data into NeoVisuals software whether that information is used to recreate an object, change how it looks, or move it. Initial Graphics Exchange Specification (IGES) and Computer Graphics Metafile (CGM) format files are discussed, as well as methods for importing unformatted raw data for visualization dataset, bar chart, proc step, sample source 7
114 Increasing the Value of Graphics Using the SAS Annotate Facility Reggie Brett Analysis of large amounts of data typically involves both summary reports and graphs. Combining the reports and the graphs can sometimes simplify the analysis of the data. The SASe system provides a method by which text strings may be combined with SAS fiJ graphs. This method is known as the Annotate facility and is an option to the GPLOT procedure. The goal of this paper is to demonstrate how, by using the SASfiJ Annotate facility. text strings and numbers may be incorporated into a SAS graph thereby increasing its value and reducing the need for multiple reports macro variable, data step, proc gplot source 4
115 Inside SAS Catalogs in Version 6 of the SAS System Rebecca Perry catalog entry, catalog, libname statement source 7
116 Institute Support for SAS Users Groups Pamela Meek,
Sally A. Roberson		 group by, sas consultant source 2
117 Integrating Graphics into a Document Using SAS/GRAPH Software and WordPerfect 5.0 Anne C. Neale,
John E. Teberg,
Brian N. Cordova		 SAS/GRAPH· graphic stream files can easily be integrated into a WordPerfect 5.0" document. An example from a document produced for the U. S. Environmental Protection Agency will be used to illustrate the procedure and the results of integrating SAS graphics. The benefits of incorporating SAS graphics into a WordPerfect document will be discussed, as will several related issues sample, phase ii source 5
118 Integrating SYSTEM 2000 Data Management Software and the SAS System: Getting Started M. Kathleen Ashlock Although the internal storage structures. of SYSTEM 2000 software and SAS data files differ, they are quite compatible in the SAS environment Interfaces are provided to ensure efficient and accurate transitions from one to the other. This paper addresses the reasons for choosing one form over the other and gives proc step, dataset, cpu time, sas log source 9
119 Integrating the SAS System with DOS Applications George F. Fraction sas output, sas log source 4
120 Interactive Random Sampling Ashok Mehta,
Paul H. Bern		 In this paper a proeed ure for developing a menu driven system of random sampling using base SAs'", SAS/STAL", and SAS/AP' is described. The system was developed on an IBM 3090, CMS, running SAS Version 5.18. A companion system was also developed for the PC using SAS-PC Version 6.03 sample, sample size, random sampling, confidence level, interval variable, proc freq source 3
121 Interfacing Normalized Relational Database Structures with SAS Software James R. Johnson,
Roger D. Cornejo		 The use of relational database (RDBMS) technology and different levels of normalization (1st, 2nd, 3rd, 4th normal data structures) is proliferating throughout the data processing industry. RDBMS systems are valued for their ability to maintain the integrity of data, reduce unnecessary data redundancy, and provide maximum flexibility in retrieval. At the same time SAS@ software is established as the general fourth generation language (4GL) tool for data analysis and reporting. Clearly, the use of relational databases and SAS software should be included in the ideal tool kit for systems development. Use of the different and distinct forms of normalized data structures, within an RDBMS, impacts the creation and use of SAS data sets. Well designed ROBMS applications typically result in normalized tables that remove repeating data groups, duplicated data, and establishes appropriate key to data associations within a table normalization, proc sql, relational database, key, sample, proc step, where statement, sas programming, dataset, create view, order by, proc sort source 8
122 Introducing New Computer Users to Release 6.06 of the SAS System Beverly Wells,
Jean Ussery		 With SASiASSIST software. getting started with the SAS System is easy, but it changes how you initiate new computer users. If you are responsible for train ing, this paper will show you how to help someone with no computer skills begin to use the SAS System with SAS/ASSIST software. What SAS!ASSIST software means for trainers? With Release 6.06. SAS Institute provides a valuable tool to help new users get started with the SAS System: SASJASSIST software. This interface between the user and the SAS System is a task-oriented, menu-driven windowing system. The primary goal of SASiASSIST software is to provide the power of the SAS System without requiring knowledge of SAS language syntax. SASi ASSIST software enables users to choose tasks they want to perform and guides them through different options. SAS code is generated based on these choices. New users of the SAS system can quickly become productive because there is no need to learn a new language: they simply pomt and select dataset, proc freq source 4
123 Introduction to CPE Using the Mainframe and PC Bruce L. Green capacity planning, sas base source 6
124 Introduction to Interlanguage Communication with the SAS/C Compiler Ingrid Ammondson A multilanguage program is composed of routines written in two or more high-level languages. Multilanguage programs are useful because they facilitate use of existing subroutine libraries migration of an application from one language to another * use of languages most applicable to a particular programming task. Writing and packaging a multilanguage program can be difficult In the IBM'" 370 environment. The SAS/E" Interianguage Communica tion feature minimizes the difficulties involved. This paper addresses the passing of data among multi language programs, with examples using COBOL, FORTRAN, and PUt. Additionally, because the link ing of multilanguage programs is an area for frequent errors, exam ples of using the ILCLINK utility are provided exception handling, autocall macro source 6
125 introduction to the C Programming Language Gary H. Merrill This is a conceptual introduction to the C programming language, concentrating on those features of C that are likely to be regarded as unusual or to be problematic for novice users of the language. The paper is oriented toward answering the question: "What infor mation will be of aid to me in learning how to program in C?H automatic variable, macro variable source 7
126 IUPUI Test Scoring System - Using the SAS System and REXX under CMS and MVS Dave Rutherford,
Shu-Yeng Wong		 proc freq, proc step source 6
127 JMP Tutoria! John P. Sall regression analysis, correlation, likelihood function, dataset, least squares mean, anova, sample, histogram source 8
128 Linear Programming Meets SAS/AF Software Danielle L. Johnson,
Martin J. Collette		 linear programming, proc step source 5
129 Loan Comparison and Analysis GuI Ege This paper describes the capabilities of the new SAS/ETS PROCEDURE LOAN. The top vote getter in the 1989 SASware Ballot for SAS/ETS was "Provide the ability to analyze variable rate mortgages." The development of PRoe LOAN was the Institute's response to this request. The LOAN procedure can analyze variable rate, as well as buydown rate, fixed rate, and balloon payment loans. It can incorporate initialization costs, downpayments, and prepayments. Furthermore, PROe LOAN offers econoII1:ic comparison of different loans at different times during loan lives interest rate, proc step source 8
130 Look Ma, No DATA Step! An Introduction to SAS/FSP Software Louise Weiler SASIFSP (Full Screen Product) procedures provide conven ient interactive facilHies for data entry and data presentation. SAS/FSP procedures allow system developers and end users alike to perform many of the same tasks that would otherwise be performed by the DATA STEP. These tasks include: Creating a SAS data set Getting data into a SAS data set EdHing data Producing customized reports This tutorial will demonstrate the FSEDIT procedure, not including Screen Control language (SCl). PROC FSBROWSE and PROC FSPRINTwili be mentioned briefly. You will learn how to perform DATA STEP tasks without writing any code. All you will need to do is respond to prompts appearing on your screen. This tutorial will also discuss the advantages and the occa sional disadvantages of using SAS/FSP proc step, dataset source 5
131 Mainframe-based PC Software Maintenance Cathy Kilburn The Education Division at SAS Institute uses personal computers for student workshops in its training centers around the country. We must assure that all the proper files are on these machines each time a student takes a class. Therefore, our goal was to develop a system to easily maintain dozens of pes from a central location, namely the mainframe computer in Cary, North Carolina. This appli cation relies on base SAS~ software on both the mainframe and the PC, especially the UPLOAD and DOWNLOAD procedures. We use the mainframe to store copies of all the PC files (in binary fonnat) along with an associated master list that identifies all the files that should be on the machines in any given training center at any given time. Then, when maintenance is run on a machine, the system uses the SAS System to upload to the mainframe a com plete directory listing of the fites currently on the PC dataset, proc download, proc step source 5
132 Making More with Less - Inventing the Three-Dimensional Pie'Chart Using Release 5.18 of SAS Software Elizabeth A. Werth macro definition, macro variable source 4
133 Management Reporting for Capacity Planning and Computer Performance Evaluation Donald T. Gantz Performance analysts and capacity planners keep management informed about how much work the system is doing, how well that work is being done, what problems exist in the system, and who has the problems. The information provided to management is backed up by a large volume of measurement data. Graphic displays are generally valuable for putting life into otherwise dull data and statistics. For computer performance evaluation (ePE) and capacity planning professionals, graphics are the critical media for communication to management. Good graphics speak for themselves. Graphics that present patterns that are readily interpretable by management also buy credibility for the analyst. The primary skill required for successful descriptive analysis and presentation is common sense, backed up by a good toolbox. SAS software is already the basic toolbox for CPE professionals. Many effective graphic techniques are easily implemented through descriptive SAS procedures capacity planning, cpu time, proc step, bar chart, trend analysis, data science, proc means source 12
134 Managing a Multi-vendor DBMS Environment with the SAS' System Tim Feetham cpu time, sample source 4
135 Managing a Multiuser SAS Database with Version 6 SAS/AF Software: Generic Tasks Todd C. Folsom permanent dataset, macro variable, quality assurance, proc step, dataset, raw data, temporary dataset source 4
136 Manipulating Data Sets and Views from Screen Control Language Ann E. Carpenter This paper discusses how data set features such as compression, Sal views, deleted observations, and WHERE clause processing affect results from certain Screen Control Language (Sel) functions in Version 6 of SASI AF" and SAS/FSP" software. The Sel functions discussed include LOCATEC. CUROBS. FETCHOBS, FETCH, A TTRN, NOTE, and POINT where statement, dataset, proc step, proc sql source 5
137 Manpower, Personnel, Training, and Qualitative Analysis System (MPTQ): A SAS Software-based Application to Satisfy DoD Requirements Aiman M. Zeid,
Robert A. Butler		 batch job, dataset, libname statement, sas batch, macro definition, macro variable source 6
138 Mastering the SAS System Using Master Indexes SAS Institute (a company) Knowing information in teday's time-squeezed, communications crammed society is 01ten less important than knowing where to find that information. Indexes, as pointers to information, have become our quick references to the world. Without them, we'd be hard pressed to locate an unfamiliar street, find a good Italian restaurant while on vacation, or even resole our shoes at the leather repair shop in the newly opened mall. For information about SAS software, you'll find everything you need in SAS Institute's four master indexes. The master indexes-to the SUGI Proceedings and to documentation for Version 5, Version 6, and the SAS/C" Compiler-are your maps to the SAS System. Used effectively, they enhance your computing power by directing you to information quickly and accurately, when you need it. Their clear, easy-to-use codes make them indispensibie guides to all SAS soft ware products proc reg, input statement source 3
139 Menu-driven Interfaces Using SAS/FSP Software and the SAS Macro Language Michael F. Perry Problem Discussed: Interfacing main frame SAS applications at sites without SAS/AF" or SAS/ASSIST" applications development software. Intended Audience: Beginning to intermediate SAS programmers who must develop applications without SAS/AF or SAS/ASSIST software. Suggested Solution: Interfacing SAS applications using SAS/FSP data entry software in combination with the SAS macro language. Benefits: Allows users to customize program/menu variables in a format that conceals application code, displays detailed user instructions and provides basic input error trapping capabilities. Outline proc step, macro variable, macro definition, macro facility, %if-%then-%else, autocall macro source 5
140 Menuing Through the SAS System: Version 6 Kirk Paul Lafler The first thing you see after beginning your SAS session is the Version 6 SAS Display Manager System, which serves as the interface between you and the SAS System. Getting around in the display manager is easy, but you may find yourself moving a hit slowly at first. This ~s quite natural since you do not know what commands and special keys are available to move you from window to window and to make your selections and entries. This paper presents Version 6 features that are useful for users and programmers. It also illustrates dataset, sas output, proc print, data science, regression analysis source 9
141 Migrating From PC DOS Systems to OS/2 Systems - Hardware Considerations David C. Brumitt i/o, proc sql source 3
142 Migrating from PC DOS Systems to OS/2 Systems - Software Considerations Julie A. Maddox source 4
143 MIS - A Decision Support System for Management Brigitta Wagner,
Kristina Hogberg		 The MIS system is a system developed for management within Marketing and Sales at Volvo Car Corporation in Gothenburg, Sweden. The project started in 1986 and MIS has become a strategic product and a powerful support for management in their decision making. We have in MIS put together information from many different areas, for example sales, stocks, car market development, finance and research and we present this information in a clear and understandable way complemented with many different graphic display possibilities. The objective for the MIS project was: providing decision makers within Marketing & Sales with "management information" on a range of key areas and key markets. The guiding principle in our work was to "create order". In the first place MIS is developed for our Management Group, but we have gradually spread the use of the system and we have today some 100 users source 2
144 Modular Course Design: Develop and Deliver a Technical Course Kim Burch problem solving, select source 5
145 Mortar and Bricks: Using Basic SAS System Procedures to Manipulate Your Data Peter Kretzman The SAS System may have had its roots in allowing statisticians to do data analysis, but that's now only one of the areas in which it excels. For most of us 000- statisticians, our principal reason for using SAS software is the ease and flexibility it provides for simple, 000- statistical data manipulation, aggregation, subsetting, and transformation. This beginning tutorial will examine a few common ways of working with some basic SAS System procedures which produce output data sets that can then be manipulated further. Detailed but generally applicable examples will show you how to work more effectively and quickly with data manipulation procedures such as the FREQ, proc step, dataset, proc freq, proc means, proc transpose, sas programmer, by group processing, proc sort, flat file, data science, data step, proc contents, set statement, proc univariate, macro facility source 8
146 Most Creative Use of the Software-Color Paula Mercer proc step, proc gmap source 6
147 Most Creative Use of the Software-Monochrome William D. Kahn dataset, proc step source 4
148 Multi-Criteria Decision Support System for Military Force Management Mary G. Crissey This application of SAS/OR software was designed for Air Staff enlisted policy planners as a decision aide in the complex arena of military force management. By coding an innovative goal programming algorithm in SAS software, the domain of Operations Research * problems which can be modelled via the SAS/OR software has been greatly expanded. No longer does the analyst need to limit his formulation of the real world into a model consisting of a set of constraints having one single objectiv~ function. By sequentially calling PROC LP and generating various specially designed reports it is possible to optimize numerous conflicting goals and cpu time, phase i source 4
149 NeoVisuals Software: Year One At The University of Guelph Anthony R. MacKay This presentation is on videotape and chronicles the developments which took place <;I.t the University of Guelph following the introduction in 1989 of the NeoVisuals 3-D modelling and animation package. Many of the colleges at the Guelph campus were quick to realise the potential of this new addition to the software library and started utilising NeoVisuals for making videos for presentations, for research and for teaching purposes. Excerpts from some of these projects are reproduced on this SUGr 15 presentation tape and an overview of some of these undertakings now follows. Initially considered a tool primarily for research, it soon became apparent that it also showed great promise as a teaching aid and this encouraged lecturers and students at the University School of Landscape Architecture to begin using it to produce animation sequences as course project assignments source 1
150 New Display Manager Features for Version 6 of the SAS System Claire Cates This paper describes the graphical and text user interface to the Version 6.06 SAS Display Manager (DM). DM is the intuitive user interface for develop· ing and executing SAS applications. The topics cov ered in this paper include 6.06 features of: a discussion of the software layering of DM a discussion of the base windows in DM * a discussion of new windowing fcatures sas output, sas log source 4
151 Nifty Tricks: Using FSCALC "Templates "for Management Reporting Rebecca A. Adams,
A. Frederick Seaman Jr.		 This paper describes some SASe techniques used in an interactive financial reporting system developed at the University of Virginia for the twenty-two cost centers in its Dining Services Division. By combininll applications of base SAS software, SAS/FSCALC", SAS/FSP", and ROSCOE/DMF" we implemented a hierarchial management reporting system that provides the following features: > Data extraction from the University's mainframe accounting system into a SAS dataset; > Interactive and batch data entry into FSCALC spreadsheet 'templates' using both base SAS software and SAS/FSP applications; > "On-line' viewing capability of operating results using ROSCOEjDMF panels and FSCALC spreadsheet 'templates'; > Management reports presented in a three-level hierarchial format by using an iterative SAS macro and SAS/FSCALC in batch mode. In creating this system. we made several discoveries about how FSCALC works, and also developed a few specialized techniques for handling our reporting requirements sas batch, dataset, proc step, macro definition, macro loop source 4
152 Nonparametric Estimates and Confidence Intervals for the Medians and Difference in Medians Between Two Treatment Groups Rodney Wong,
Reiling Lee		 dataset, confidence interval, confidence level, sample, proc step, proc rank, wilcoxon signed-rank, rank test source 5
153 Nonsymmetric Bootstrap Confidence Intervals Using SAS/IML Software on a PC George W. Pasdirtz In this paper, PC SAS/IML modules are developed which calculate percentile method confidence intervals and vanants. The modules will work on distributions of parameters generated by any Monte Carlo technique, Including the bootstrap bootstrap, confidence interval, sample, proc calis, monte carlo, proc step, normal distribution, probability distribution, regression analysis, confidence level source 4
154 Organizing CMS Files Using the SORT Procedure, REXX, and FILELIST Kimberly J. LeBouton You can make many changes to this simple REXX XEDIT macro to make it more useful for your particular environment. For example, our installation allows users to pack large files. Since it is feasible that the file produced by FILESAVE can get quite large, it may be necessary to pack the file. A modified version of FILESAVE has been written which searches for a packed file. If a packed file is found, you are' informed to unpack the file and rerun FILESAVE. Cerlain variables from FILEUST were chosen to be included in' the FILESAVE output (i.e.: FILENAME, file type, file mode, block size, and date). Other variables are also available with FILEUST and may be more important to you--e.g.: format, line record length, records, and time. A simple modification to the SAS program allows you to extract the information you need. Just remember to leave enough space to add an adequate file description merge statement, proc sort source 5
155 Overview of SAS/GRAPH Procedures and Facilities Vincent P. Tillman proc step, proc gchart, bar chart, proc gplot, proc freq, proc greplay source 6
156 Overview of the SET, UPDATE, AND MERGE Statements Richard Alonso missing value, by group processing, proc step, proc sort, proc print, set statement source 5
157 Parameter Estimation For Models Defined By Differential Equations James E. Dunn,
Kumiko Smith		 macro definition, proc step source 10
158 Performance Considerations for the Database Engines Howard Plemmons Jr. This paper identifies some of the issues to consider when accessing your Data Base Management System (DBMS) through the SAS· Version 6 database engines. Both SAS and DBMS performance Issues are identified throughout the paper. The focus of this paper is on the current Version 6.06 engines: SYSTEM 2000: DB2; SaUDS; RdblVMS' and ORACLe proc print, proc step, where statement, dataset, cpu time, proc sql source 6
159 Performance of Multiple Comparisons in Repeated Measures Designs under Nonsphericity Steven F. Thomson repeated measures, covariance, anova, type i error, correlation, proc step, proc anova, sample source 6
160 PERSPECT: Extracting Perspectives from Super-Encoded Data Vincent P. Tillman regression analysis, proc format source 5
161 Planned ESA/378'"Exploitation by the SAS System under MVS Keefe Hayes IBM·S· powerful Enterprise Systems Archltecture/370- provides many new capabilities for system and a!'P'ication software. The exploitation of these capabilities in future releases of Version 6 of the SAS- System under MVS will offer enhanced function and per formance dataset, reference data source 4
162 PLOTCORR: A SAS Macro for Appending Correlation and Regression Statistics to the PLOT or GPLOT Procedures Erik J. Bergstralh regression analysis, correlation, scatter plot, proc step source 4
163 Polychlorinated Biphenyls Information Management and Modeling System for the Unsaturated Soil Zone Jing-Yea Yang,
Shi-Tao Yeh,
Frederick Bopp III		 sensitivity analysis, elapsed time source 6
164 Porting C to the IBM Mainframe Environment Gary H. Merrill ThiS paper deals with issues encountered in porting existing C pro grams to the mainframe environment Following a discussion con cerning what it means to say that a program is portable, this paper emphasizes program features that are likely to inhibit portability to and from the mainframe txt, character constant source 5
165 Processing of External Data Sets in Version 6 under MVS Amitava Ghosh,
William F. Heffner,
Hugh D. Landis		 This paper focuses on the available external file capabilities of Release 6.06 of the SAS" System. This paper emphasizes the fol· lowing new and improved features: dynamic allocation through the FILENAME statement concatenated data sets PDS considerations sequential use of POS members ability to aocess VSAM data sets ability to update ISPF statistics This paper discusses these features by defining external files to the SAS System with the FILENAME statement and using external files in the SAS DATA STEP through the INFILE and FILE statements. Fur thermore, this paper addresses additional external file features, including accessing VSAM files and VTOCs, MVS windows, the FILENAME= option, and the FILEVAR= option sas log, file statement, filename statement, put statement, infile statement, input statement, dataset source 8
166 Program to Generate Atkinson's and Resistant Envelopes for Normal Probability Plots of Regression Residuals Rafael Flores,
Virginia F. Flack		 regression analysis, cook's distance source 8
167 Programming Techniques Using Choice Groups with Extended Tables in Version 6 SAS/AP Software Richard A. Ragland The use of choice groups with extended tables is a new feature of Version 6.06 SAS/AF software. A choice group is a list of mutually exclusive items that users may select. Extended tables are dynam~ ie tables that display information. This paper will discuss program ming techniques for generating dynamic selection lists to combine the features of choice groups with the features of extended tables. The discussion will include a more extensive definition of choice groups and extended tables, an explanation of how to set up choice groups and extended tables in a SAS/AF PROGRAM entry, and examples of Screen Control Language (Sel) programs that are easy to maintain and make applications more user-friendly. Definition of Choice Groups A choice group is a new feature of Version 6.06 SAS/AF soft\ovare, and is an attribute of SAS/AF PROGRAM entries. A choice group is a set of fields that are similar in nature, and are referenced in SCL by a single choice group name dataset, where statement source 6
168 Programming with SAS Functions, with a Special Emphasis on SAS Dates and Times Ann K. Doty time variable, date variable source 4
169 Projecting Course Enrollments at Syracuse University Using the SAS System Mark A. Storrings Personnel in the registrar's office at Syracuse University needed a method of informing the schools and colleges within the University as to the demand for courses during the pre registration process. This paper will show how the SAS'" System was used to project final course enrollments so that academic units on campus could make informed decisions regarding the allocation of resources for additional sections of courses in high demand. This session will highlight: 1) efficient dataset, subsetting if source 4
170 Quick Start to C Using the SAS/C Mainframe Compiler Dee Stribling This paper suggests ways you can use SASIC'" documentation to enhance SASrC program development. The paper begins with an overview and orientation to the documentation. After you're familiar with what's available, you'll follow the development of a C program in tenns of .that documentation. The example program provides the basis for discussing how and where many SASIC features, libraries, options, and so forth used in the program's development are docu mented. Knowing your way around the documentation can help you take advantage of features such as the optimizer and debugger, resulting in more efficient SASIC applications. This paper is directed towards new users of the SASIC Compiler under either MVS or CMS. It is assumed that you have a basic understanding of the C language at the application programming level. The focus is on the documentation accompanying the SASIC Compiler and the compiler rather than the C language itself libname statement, sysrc source 8
171 Real Men Do Use Menus: Developing Applications with SAS/AF Software Kenneth E. Johnson macro variable, proc step, character variable, macro definition, dataset, numeric variable, sample source 9
172 Reproduction of Ambulance Runs of the Chicago Fire Department John Barry The Reproduction of Ambulance Runs Program can assist decision making in the Chicago Fire Department by suggesting more efficient deployments of present am bulance companies and optimal deployment of new com panies as they become available. The output from the program pinpoints the areas of the city with the most demand for emergency medical services and areas where all incidents may not be responded to within five minutes. The program can test various locations of new ambulance companies before deployment, and examine the effect on coverage throughout the city. The areas that are under covered. that is. not within five minutes of an ambulance company are also isolated in the output report proc freq, proc step source 3
173 Resampling-Based Multiple Testing Peter H. Westfall,
Youling Lin,
S. Stanley Young		 proc step, interval variable, bootstrap, correlation, multiplicity, rank test, resampling source 6
174 Rod Mill Billet Surface Quality Reporting System John A. McNamara,
L. Richard Woodyatt		 options statement, data cleaning, sample, dataset source 8
175 SAS Commands from the CMS Command Line Chapman P. Gleason This paper describes methods on how Systems Application Architecture REXX can be used to develop powerful SAS procedures from the eMS Command line. A REXX exec called SASPROC will be described in detail showing how you can extend the SAS language to the CMS environment. Introduotion In a CMS environment a SAS programmer frequently wishes to FSEDIT/FSBROWSE or otherwise process the contents of a SAS data set or SAS full screen catalogue. The usual way of doing this is rather tedious, you either: 1) Use XEDIT and key in a filedef and PROC FSEDIT into a SAS source code file or 2) enter an interactive SAS session in Display Manager (DM) and key in the filedef and PROC FSEDIT statement. An experienced programmer sees this tas~ as exceedingly redundant, borlng and easy to cure with some programming in REXX proc step, dataset source 4
176 SAS Institute's Online Customer Support Facility Alissa Schleich SAS Institute's Online Customer Support Facility is a free bulletin board·type system that the Institute uses to provide 24-hour com munication with customers. The support facility's main purpose is to provide technical support. The following technical support-related services are provided: ability to track problems access to aU production zaps and corresponding Usage Notes access to problem alert letters ability to upload and download files associated with outstanding problems or questions. Other services that are not directly related to technical support are ability to view training course schedules ability to order publications * ability to browse Institute-related announcements. In addition to a general overview of the support facility, this paper gives several tips that will help SAS" Software Representatives and SAS Software Consultants effectively use the Online Customer Support Facility to provide on-site support to SAS users source 3
177 SAS Macros for Source. Program Tracking John Henry King,
Donna S. Goldstein		 This paper describes a macro system designed to reduce confusion over the origin of output. It is primarily designed for eMS SAS users and for jobs submitted to OS from eMS. Using the macro system, output can be annotated so that one can easify determine the origin of the source program producing the output, and the date and time it was run. Specifically, the macro traces the eMS Userid. the eMS minidisk file that created the output, and the date and time the output was produced. Output from SAS procedures. PUT statements, and SAS/GRAPH8 can be annotated using this system composed aftwo EXECs and a SAS macro. To initiate the source program tracking system, users must modify their SAS "EXEC and write a QUERYCON EXEC to communicate with VMBATCH0. Changes to the EXECs are made once, and saved permanently, on the user',s minidisk for future use macro variable, keyword parameter, %macro parameters, macro call, macro conditional, pointer control, put statement, macro loop, character constant, macro definition source 3
178 SAS Software and the Single Research Scientist Michael S. Lajiness proc step, flat file, proc printto source 6
179 SAS Software Performance in the Persona! Computer Environment Donald R. Deese,
John D. Ebner		 The SAS Institute has ported a number of SAS software products to the personal computer environment. The SAS System Version 6 for Personal computers potentially offers an inexpensive and convenient interactive environment at the workstation level. However, the potential advantages of the SAS System for Personal Computers can be achieved only if you have a proper microcomputer configuration, use proper supporting software, and select proper SAS configuration options. This paper describes benchmark results illustrating the performance improvements that can be obtained using proper configurations, supporting software, and SAS configuration options source 10
180 SAS System Performance on 80386 Class PCs Mark Pabst With the advent of the SAS· System on microcomputers, some confusion has arisen about the best hardware and software for maximum performance. This paper discusses in depth two 80386 systems. The first is a 16·MHz machine with two 28-ms hard drives; the second, a 25-MHz machine with a 14-ms hard drive. Several other systems are included for comparison. Variations on the following setups are considered - math coprocessor I CPU speed, EMS, hard drive utilization, operating environment (DOS· 3.21. DOS 3.3 and VMj386), and disk caching. Typically, users plagued with poor system performance try to solve the problem by purchasing a faster PC with more memory. In most instances, this will not have the greatest impact on system performance proc step, sas function source 6
181 SAS/ACCESS Interface to ORACLE and Rdb/VMS;"Release 5.18 and Release 6.06 of the SAS System Margaret Underberg The transparent access of data from external databases is now a reality under Version 6 of the SAS" System. Multiple engine con cepts in the Multiple Vendor Architecture (MVA) bring increased flexibility and power to SAS users across many of the SAS plat forms. Under the VMS- operating system, two database engines have been developed for Release 6.06 of the SAS System. The SAS/ACCESS~ interface to ORACLE" and the SASfACCESS inter face to RdbNMS- enable you to combine the power of the SAS System with your data stored in these database management systems (DBMS) proc step, dataset, select, oracle access, proc sql, where statement, order by, proc print source 5
182 SAS/ASSIST Software under Release 6.06 Katherine J. Roggenkamp,
Gail Kramer,
K. Deva Kumar		 SAS/ ASSIST software provides a menu-driven, task-oriented interface to the most commonly used features of the SAS Sys tem. This interface makes it simple to perform tasks in the areas of data access, data management, data analysis, and data pre sentation. SAS/ASSIST software is designed to serve the needs of users with many different levels of expertise. SAS/ASSIST software was constructed using tools available to any user of the SAS System. Thus, the software can be cus tomized to meet the needs of a particular site. Tools to aid in application development are included as part of SAS/ASSIST software macro variable, libname statement source 9
183 SAS/GRAPH Software and Numerical Taxonomy Dan Jacobs A goal of numerical taxonomy is to group organisms based on physiological data, RNA sequences, and results from DNA homol ogyexperiments. The results of cluster and principal component analysesaretypicallydisplayedindendrograms, similarity triangles, and principal coordinate plots. These graphs can assist in the interpretation of the results and in the delineation of the organisms into possible taxonomic groupings. The results of two cluster analyses using different sets of data or different cluster methodolo gies can also be compared graphically. SAS<3 and SAS/GRAPI-P software have a number of procedures which perform the data analyses and produce graphs (e.g., PROC CLUSTER, PROC GPLOT, PROC PRINCOMP and PROC TREE). While principal component plots are easily produced using PROe PRINCOMP, PROC GPLOT and PROC G3D, dendrograms produced by PROC TREE are both hard to interpret and not of publication quality proc step, proc princomp, proc cluster, cluster analysis, data step, proc gplot, proc tree, confidence interval, raw data source 6
184 SAS/GRAPH Software Meets the Logo Turtle Michael Friendly This talk describes a set of SAS macros which provide the ability to draw with the SAS/GRAPH ANNOTATE facility using the "turtle~relatrve" drawing commands of the Logo language, rather than absolute X·Y coordinates. This makes drawing much easier for many types of figures. Another key feature of Logo is recursion, which also simplifies many graphics procedures, Since the SAS DATA STEP has neither local variables, nor recursion, it is hard to some things that are very easy in Logo. However, SAS macro variables are local variables, and recursive macros which model Logo procedures can be defined. A variety of Logo-like designs carried out with SAS/G RAPH are illustrated macro definition, proc step, data step, macro variable source 6
185 SAS/GRAPH Software to Macintosh : A User's Experience on - a VAX "System Leanne Whiteside SASIGRAPH is an excellent tool to graphically summarize data. However, customizing a graph (positioning labels, adjusting spacing. etc) can be tedious and difficult. One option in SAS version 5.18 is to use the ANNOTATE = utility. but even an experienced user can find it difficult. An alternative is to generate the basic graphics code on the VAX using SASiGRAPH, then download this code to the Macintosh. The graphics code can then be converted to a format compatible with a Macintosh graphics application. This conversion is accomplished with an application developed by SAS Institute, Inc. called META. Once the graph is on the Macintosh, customization is a matter of point and click. The graph can be incorporated into final papers and printed easily on laser printers txt, proc step source 4
186 Scheduling Programming Projects Using SAS/OR Software in Version 6 of the SAS System Eric C. Brinsfield SAS/OR" software provides several facilnies for managing projects and resources. This tutorial describes how to design a menu-driven system for managjng muniple tasks that are to be completed by several differeni individuals. Using the CPM procedure, the system tracks utilization of manpower, monnors project progress, and estimates feasibility of starting new projects wnhin a given time frame. The system uses Release 6.06 Screen Control Language wnh SAS/FSP" and SAS/AF" software to make entry of the projects simple proc step, where statement, macro variable source 7
187 Scientific Workstations and SAS Software: Getting Jobs Completed Robert W. Young SAS software is an integral tool within the scientific community in the Mid-Pacific Region of the Bureau of Reclamation. Principle investigators. analysts, and program managers are able to assess the Region's hydrology, water quality, and biological conditions effectively with SAS statistical and graphic software. The Regional Staff are using "workstations" to process data and information. Key elements of the stand alone units include communications, data access, graphics, statistics, modeling. and report writing. Paper focuses on the role and value of each component. The units include 386 class microcomputers as well as UNIX' based systems. Mainframe computers are also being used and include the Bureau's VAX" 8300, and E.P.A.'s I.B.M. computer system in North Carolina. SAS software is available on each platform. Distributive data processing has proved to be cost effective in both time and money regression analysis, data cleaning source 4
188 Screen Control Language - An Invaluable Tool in Applications Development David S. Septoff Screen Control Language (Sel) is a programing language that pro vides functions and routines to control fields on windows and screens to manage the windows and screens that make up an appli cation. With Sel, the SAS" programmer can perform real-time cross-validation of variables control the application and interact with the SAS System control access to, and obtain information from, SAS data libraries control communication in an application. * control error handling and provide messages specific to the application This tutorial demonstrates that Sel provides the power and flexibil ity needed to build all types of applications, from simple programs that accomplish a few tasks to sophisticated SAS systems dataset, data cleaning source 4
189 Segmenting the Tourist Market Using Factor Analysis and Cluster Analysis Marcus Schmidt,
Hans Merser		 During recent years life-style segmen tation (psychographies) has been a re search topic of increasing interest in developing appropriate marketing strate gies and target marketing. In a research project, sponsored by the Danish Tourist Council, a sample of 4,000 respondents was interviewed with regard to their vacation and tourist behavior, attitudes and preferences. The data from the section with psycho graphic variables were factor analyzed. preliminary cluster analysis; of the raw data matrix (observations x factor scores) indicated that the data structure might be plagued with outliers. Then another cluster analysis was con ducted (Anderberg;s nearest centroid sorting) with the aim of detecting out liers. It was suggested that an ; optimal; or ;true; clustering would happen when using 4 clusters. Finally the clusters were interpreted with regard to their socio-demographic behavior and psychographic attitudes, and the outlier, cluster analysis, factor analysis, proc step, proc princomp, sample, missing value, scree plot source 6
190 Selecting Optimal Branch Locations Based on Geographic Distribution of Service Call Volume Robert P. Wederich Jr. The ability to respond quickly to customer needs is a key element to the success of any service-oriented organization. Given that a good network of roads exist, selecting optimal branch locations to minimize physical distance between a service organization and its customers will benefit both the customer and the service provider. The present paper employs SAS/GRAPH* to examine the geographical relationship between customer service branch locations of Digital Equipment Corporation (DEC), and the distribution of service call volume within the northern half of the State of New Jersey. A visual presentation of the data aided greatly in detecting less than optimal branch locations I and in helping upper management understand both the problem l and some proposed solutions source 5
191 Shortcuts to Customize Clinical Trial Reports Aileen L. Tsien Accuracy, efficiency and flexibility are essential to research in all fields. This paper illustrates ways to customize clinical trial reports with little programming or editing efforts as well as to automate the production of a series of reports. Although written in the context of clinical research, the general approach to tailoring and mechanizing report production described in this paper is applicable also to other research disciplines proc step, proc format, p-value, clinical research, format statement, clinical trial, macro variable, proc genmod, proc glm source 6
192 SIGNIF, a C Function to Compute Significant Digits B. M. Horwedel,
J. E. Horwedel		 This poster presents a C function, SIGN IF, that can be called from a SAS program to return a value rounded to a specHied number of signHicant digns. The rules for rounding a number to n significant digns as implemented by SIGN IF are: (1) Truncate the number to n digns, and treat the excess digns as a deCimal fraction; (2) If the fraction is greater than 1/2, increment the least significant digit; (3) If the fraction is less than 1/2, do not increment; (4) If the fraction equals 1/2, increment the least significant digit only H n is odd numeric functions, sas function source 3
193 Software Tools for Processing U.S. Census Bureau TIGER Line Files John G. Blodgett,
John McIntyre		 The U.S. Census Bureau is generating block level detail computer ized maps as a part of the 1990 Census of Population and Housing. The availabitity of such computerized files will greatly enhance graphical display of demographic data in govemment and bUSiness applications. Prototypes of these maps are available in line segment form. SAS" macros are presented that convert these line segments into geographic polygons for mapping. Files are converted in a main frame session, and resultant SAS files are downloaded using the micro-to-host link, for polygon chaining and SAS/GRAPW process ing proc step, proc gmap, sample source 5
194 Solving 2,000 SAS Problems a Year. Help Desk Personnel - Solutions Faster Than a Speeding SAS User Robert G. Hall,
Catherine Loveless-Schmitt		 This paper discusses operating a successful Help Desk supporting over 500 active SAS users. We support users working on systems ranging from pes and Unix workstations to VAX mainframes. In our approach the help desk personnel try to identify the type of user and the nature of the problem quickly and then tailor the response to the user. Responses range from brief explanations of SAS or operating system syntax to extended systems analysis. The paper also discusses additional activities that may be conducted during those periods when there are no specific requests for assistance. These activities are designed to improve the general level of SAS support by developing generalized solutions to problems and prototype systems using different SAS techniques and products. A SAS help desk assists users in getting their SAS code to do what they want SAS to do. Our most recent experience with help desks involves our work at the Census Bureau proc step, data wrangling source 3
195 Solving Business Needs as a Team Roger L. Crouse sample, data governance source 6
196 Solving Grouping Problems: Using the POINT= Option Dalia C. Kahane data step, conditional do loop, dataset, if-then-else source 6
197 Sponsoring SAS Training through a Local Users Group Tom Krasner This paper describes the experiences of the South Florida SAS Users Group in sponsoring a SAS Basics course and a SAS Macro Language course. The users group organized and marketed the courses. The Education, Marketing and Sales Department of SAS Institute provided the speakers and course materials, collected course fees, and paid course expenses. For the user community the courses allowed individuals to obtain training they otherwise would not have received. The two courses targeted different sections of the user community: the SAS Basics course for novice programmers; the SAS Macro Language for intermediate to advanced programmers. For the users group the courses supplied the opportunity to provide members a visible l valuable product which they could use to justify par ticipation in the users group. It also helped sell the benefits . o~ SAS* software and to develop a const~tuency of SAS users within the attendees' companies macro definition, macro facility, sas base source 4
198 Standards and Macros: The Keystroke Busters! Judy A. D. Alicante As SAS programmers, how many times do you have to key in: PROC CONTENTS DATA=DDNAMEo_ALL_ NODS; RUN; How about proc contents, data wrangling source 3
199 StatLab Revenue System Pamela G. Spurrier dataset, macro definition source 4
200 Strategies for Planning and implementing the Conversion of SAS Applications from Version 5 to Version 6 Harriet Janes,
Sarah Dietrich		 This paper discusses what you should be dOing and thinking about as you develop and maintain Version 5 applications so your installa tion can convert to Version 6 with a minimum amount of resources and problems. It is directed toward managers who will be adminis tering this conversion and technicians who will be responsible for implementing the conversion. The use of SAS· tools that can sim plify this conversion process, induding the SAS'" macro and autocall facilities, and SASJDMI software is also discussed libname statement, dataset, proc step, options statement, autocall macro, macro definition, sas base, macro variable source 11
201 Subsetting and Combining SAS Map Data Sets to Produce Custom Maps Susan J. Slaughter source 6
202 SUN_Q: Measuring Customer Delight Using SAS and Sybase Software Claire Casteil,
Paul Desjardins		 This paper presents the conversion of a time-consuming manual reporting system to an on-line data collection system with automated reporting and graphics. The project is a symbiosis of the SAS System's statistical and reporting tools with the Sybase database, all in a SunOS environment. The system, "SUN_Q", collects the results of a monthly customer satisfaction telephone survey (recently dubbed "customer delight" by our 35 year old CEO). The complex ities include: Downloading selected survey sites from an IDMS database on a mainframe. Collecting the data in Sybase and passing it to the SAS system. ... Passing off serious customer issues to a sub-system for personalized handling,which would automatically generate E-mail and collect responses. Making the system modular, enabling other company divisions or geographies to incorporate only the needed pieces proc step, proc sql source 6
203 Table Lookup without a Table Joseph A. Damico lookup, proc step, proc format, put statement, substr function, dataset, character variable source 4
204 Teaching the SAS Programming Language to Programmers and Non-Programmers Wendy London proc step, sas programming, sas programmer source 6
205 The Analysis of Data from 2 X 2 Crossover Trials with Baseline Measurements Joe DiGennaro,
William C. Huster		 crossover study, treatment effect, covariance, baseline, washout period, covariance structure, treatment period, proc genmod, proc step source 6
206 The Bootstrap and Jackknife Estimates of Variance: Two Resampling Techniques Using Release 6.03 of SAS/IML Software Margaret G. Hirsch,
Tamara R. Fischell		 bootstrap, sample, probability distribution, macro definition source 5
207 The Calculation of Average Run Length for a Nonparametric CUSUM Procedure Dennis W. King,
Michael T. Longnecker		 A method of calculating the probability mass function of the Wilcoxon signed rank statistic is discussed. The algorithm pre sented, written using PROC IML, is based on the work of Milton (1970). A brief sum mary of a newly developed error bound for this algorithm is given. For process control schemes based on the Wilcoxon signed rank statistic, the cal culation of the Awrage Run Length (ARL) of the scheme requires the evaluation of the probability mass function of the Wilcoxon statistic under a shift in the location param eter. The application of the algorithm's out put to the calculation of ARL is shown in de tail sample, proc iml source 7
208 The Consultant's Life Is/Is Not A Happy One Tom Marx source 3
209 The Development of an Integrated Tracking Application in an Information Systems Environment Julianne Staub Perry Tracking problems that arise in an Infannation Systems environ ment is not a simple task but is very necessary to ensure a quality operation. Problems that should be tracked in Information Systems arise not only in the data center itself, but also in the areas of hard ware communication and management infonnation systems (which includes user services). All three areas need special attention, and though they differ in many ways, the basic tracking information is shared among them. Because of this, it is optimal to create a system that combines these areas into one Information Tracking System. Data are shared among each, yet they also contain information unique to their particular environment. It is also optimal to provide a way for incidents to be routed from one area to another without leaving the data set. Accomplishing these tasks involves the use of the powerful SAS" System. An example includes using SASfFSP 8 to display the same SAS data set in three different ways dataset, master data management, management reporting, proc step source 8
210 The Distribution of Cronbach's Coefficient Alpha from the CORR Procedure of SAS Software Andrew J. Cucchiara,
Susan J. Kenny,
J. Paul Costiloe		 sample, sample size, correlation, dataset, null hypothesis, proc corr, monte carlo source 6
211 The Distribution of the SAS System for PCs via a Mainframe Computer William G. LeBlanc sample, txt source 6
212 The GLIM Procedure: An Interface to the SAS System P. M. Grambsch,
Jon L. Kosanke,
Terry M. Therneau,
D. J. Schaid,
R. Zinsmeister,
H.S. Wieand,
Kenneth P. Offord,
J. J. LarsonKeller		 covariance, estimate, binomial distribution, dataset, class variable, proc step, degrees of freedom, predictor variable, poisson distribution, by group processing, response variable, variable list, probability distribution source 5
213 The Hazardous Exposure Computer Control System Richard P. Brodine The Occupational Safety and Health Act of 1971 focused attention on the possible overexposure of Gulf Oil's refinery workers to hazardous substances. A project team was formed of employees from Gulf Management Sciences, Gulf Research & Development Company's Industrial Hygiene Group and the Information Sciences Division to investigate the matter and develop a statistical information system capable of determining the levels of hazardous substances in the refinery environment. Hazardous Substances at the Refinery Major oil companies are in the bu~iness of refining gasoline from crude 011 and employ thousands of persons to operate and manage the refining facilities proc step, anova source 6
214 The Impact of Machine Architecture on the Implementation of the SAS System Larry J. Noe missing value, proc genmod, proc glm source 6
215 The Painless Path to Release 6.06 of the SAS System Robina G. Thornton,
John C. Boling		 This paper addresses two main issues. The required changes for application conversion from Version 5 of the SAS" System to Release 6.06 is addressed first. Next, selected new features avail able in Release 6.06 of the SAS System that are not required for application conversion but can enhance the application in terms of efficiency, maintainability, or required development time are dis cussed proc step, dataset, proc sort, sas base, sas library, missing value, libname statement, options statement, proc sql source 6
216 The Patient Discharge Reporting System Kathleen A. Jablonski proc step, macro variable source 3
217 The QUERY Window Paul Kent This talk describes work in-progress on an interactive Sal Query building environment for the SAS system. It uses full screen menus and point-and-shoot concepts for selecting tables and variables as it leads you through the construction of a query. The constructed query can be run to preview the data selected, saved as program text for incluSion in other jobs, or saved as an SOL View fa use at a later time by other SAS procedures source 4
218 The Relational versus Hierarchical Database Debate Continues Alma Zuniga The marketplace of database products has been shaped in large part by the leading academicians in computer science and, of course, by IBM. Once the relational technology came on the scene, the leading scholars were quick to embrace the idea. After all. hardware storage was getting cheaper and CPU power. once isolated in massive computer rooms, was making its way to the desktop_ Is todays technology leading towards a relational model? The purpose of this paper is to explain the difference between a relational and hierarchical'data structure and present a description of the characteristics that define these structures. Although the data structure is a part of the database design, there should be other considerations when selecting the correct database management system relational database, key, dataset source 4
219 The REPORT Procedure: Filling the Gap in Report Writing Johnny B. Andrews,
Lynn Hannah Patrick		 With Version 6 SAS" software available to all platfonns, the REPORT procedure will be the first choice in report writing needs for all SAS programmers. This tutorial emphasizes the functionality of PROe REPORT, by comparing and contrasting its report writing capabilities with those available in the PRINT, QPRINT, and TABULATE procedures and the SAS DATA STEP proc step, proc report, proc print, proc tabulate, libname statement source 7
220 The SAS Documentation Library for Release 6.06 Software John H. Gough,
Sean M. Gargan		 Release 6.06 of the SAS'" System marks the beginning of a new era for SAS documentation as well as for SAS software. SAS Institute started with a philosophy stating that books are not just infonnation, they are productivity tools. Finding that SAS users are diverse in their needs, we developed a new library architecture, one that pro vides two types of book for learning tasks and one type for looking up features. As the library has grown, the task of selecting documentation has grown more complex. New services are planned that will make this task simpler. For the future, we intend to continue to improve the quality of documentation by writing new books, involving users in evaluating them, and exploring new technology for online access to infonnation sas base, options statement source 6
221 The SAS Macro: An Aid to the User Robert E. Johnson This paper is presented as a beginning tutorial on the SAS" macro facility. If you write your own code, either simple or complex then you may find the macro facility nseful. The purpose here is to show you some of the fundamentals, share some examples, and start you on the path of learning more about the macro facility. It should be understood that this tutorial does not present an exhaustive coverage of macros. Indeed, to fully appreciate and utilize the power of macros, one should have mastered the art of structured progrannning. But the rest of us can still find useful applications. Version 5, 5.16 or later, of the SAS" System allows full use of macros under the CMS, as, VM/PC, and VSE operating systems. You will have limited use under the VMS operating system. Version 6.06 should bring VMS into the fold. The macros presented here were tested on version 5.18, version 6.03 (for the PC), and version 6.06 (beta) macro variable, macro definition, data step, keyword parameter, %macro parameters, call symput, dataset, do loop, macro facility, title statement, macro loop, macro parameter, anova, proc means, %if-%then-%else, proc step source 7
222 The SAS Multiuser interactive System in Version 6 of the SAS System Tony Valmassoi The SAS· Multiuser InteractiVe System in Version 6 of the SAS Sy~ tern allows an alternative to running the SAS system under ICCF. This approach enables the SAS System fuming in a single region to service multiple terminal users, workstation users, or both. Bene· fits include better memory utilization, increased capacity for interac tive users, and faster response time. This paper addresses the design considerations of the SAS Multi user Interactive System. Induded are discussions of the manage ment of multiple users, the VT AM- Interface, asynchronous processing. the relationship to CICS, and extensibility issues sas base, sas batch source 8
223 The SAS Supervisor - A Version 6 Update Donald J. Henderson,
Merry G. Rabb,
Jeffrey A. Polzin		 This tutorial-updates-a talk given at previous SUGis that discussed the functions of the SAS"'supervisorduring the compilation-and -exe -eution of aSAS OATA step program. Under Version -6 of the SAS System, the SAS supervisor provides interface -5efVices at a higher level than before, and its primary actiVity in this oontext is to deter mine whether a DATA or PROe step is being invoked. Once invoked *. the DATA STEP controls its own environment during-both compile and,executiorr-phases, This-tutoriatpresents alcgical model describiflg-both--the--compile-and -execution time actions of-the DATA STEP processor by group processing, dataset, set statement, merge statement, missing value, update statement, where statement, sample, conditional do loop, output, do loop, keep statement, proc step source 9
224 The SAS System - A Full-Featured Utility and Language for Programmers Steven J. First SAS software is certainly well known as an end user tool for the more casual computer user. Data processing staffs however can exploit SAS as a full featured programming language to produce both one shot applications and full production systems. This paper will concentrate on the features of SAS that make it an excellent tool for programmers. It will present a series of short applications that show how SAS can fit in with the other languages used in your shop and make the programmer's job easier proc step, proc freq, proc rank, data step, sample, proc sort, batch job, flat file source 10
225 The SAS System as a Database Manager in Environmental Problems Ronald R. MacDonald,
Anthony B. Kreamer		 Work on environmental problems has become so complex that computerized methodologies are needed to meet the environmental objectives of all studies. This system must be powerful, with a broad range in analysis functions, yet be an II user friendlyll system. SAS was the choice of the application language for the development of this system by Roy F. Weston, Inc. (WESTON). This paper describes the evolution of the use of SAS as a database management system for environmental problems into one integrated system using SAS Release 6.03 products. A detailed description is given describing a menu driven system developed at WESTON using and integrating SAS macros and the following PC SAS Release 6.03: SASjAF, SASjBASE, SASjFSP, SASjGRAPH, and SASjSTAT. This application combines database management, data entry, statistical analysis, reporting tools, and high resolution graphics into one integrated system sas base, macro definition source 5
226 The SAS System as Simulation Language: An Application to Tunnel Construction Modeling Ronald R. MacDonald,
Shi-Tao Yeh,
Guillermo F. Salazar		 univariate analysis, random number generation source 6
227 The SAS System Earns an MBA: Using SAS Software in an MBA Economics Course Joseph E. Earley proc step, proc reg source 2
228 The SAS System for PCs: A Time Saver from Spreadsheet to Mainframe Jane C. Khoury,
Philip R. Khoury,
Vicki S. Hertzberg,
Chih-Yu Chen,
Richardus Ross		 The spreadsheet format for data collection mirrors the thought processes of the basic scientist and thus its use is prolHerating in the laboratory setting. This has given laboratory personnel a great degree of 'control' over their data. The difficulties now posed to the statistician are to find a reliable, efficient way to translate data from the spreadsheet format into an analytical format. The specific problem posed to us involved taking data of a longitudinal nature from the laboratory to analysis. Data are currently entered into LOTUS 1·2·3 but need to be available on the mainframe for analysis by SAS and BMDP. Automation of this procedure has been made possible by initially invoking the TRANSLATE facility in LOTUS 1-2-3 and creation of a DIF file. Submission of a SAS batch procedure involves preprocessing of the DIF file then invoking PROC DIF. Using a custom convert batch utility described in SUGI t 4 Conference Proceedings the SAS file can be uploaded to the mainframe proc step, dataset source 4
229 The SAS System for PCs: Management Reporting Using Data from External Sources Hugh G. McCabe This paper describes the use of the SAS System for PCs to transfer data from external sources and do management reporting. Discussion will focus on the development of a decision support system in a large health maintenance organization. Areas covered will include-general experiences and problems concerning the collection, processing and dissemination of information insurance claims, dataset source 3
230 The SAS System for PCs: Some Speed Surprises Richard H. Browne proc step, proc sort, proc means, regression analysis, proc univariate source 10
231 The SAS System for Statistical Graphics - A Preview Michael Friendly This talk presents an overview of methods for statistical graphics covered in my forthcoming book. The SAS System for Statistical Graphics. In particular, I i1lustrate the design and implementation of custom graphic displays for: symmetry transformation plots scatterplots, enhanced with marginal boxplots and concentration ellipses. display of additive fits and residuals for two-way tables. scatterplot matrix for multivariate data box plot, proc step, regression analysis, proc gplot, correlation, outlier, macro definition, macro loop source 6
232 The SAS System on AIX "(OSF ") Platforms Michael K. Burns This paper is an overview of the implementation of the SAS System under AIX'", which is IBM Corporation's implementation of the UNIX operating system and is the basis for Open Software Foundation's OSF/l'" operating system. My understanding of the history and direction of AIX development is presented. The features of Release 6.07 of the SAS System and AIX that provide interoperability between AIX platforms and other UNIX and non-UNIX machines is reviewed catalog, proc download source 5
233 The Secret Life of the SAS DATA Step Thomas A. Miron When you use the SAS DATA STEP you are writing a computer program just as sure as if you were using a language like COBOL, BASIC, or C. SAS provides many features that make DATA STEP programming much easier than programming in one of the traditional languages. But, sometimes these features mask the fact that you are writing a program at all! An understanding of how the DATA STEP works, Il un der the covers u, is essential for attacking complex programming tasks and debugging. This tutorial outlines the special features and hidden functions of the DATA STEP and common DATA STEP statements in an effort to enhance this understanding. While conducting SAS training I have often run into students who have used SAS for years without realizing or fully appreciating how the SAS system and the SAS DATA STEP in particular really works. The distinction between DATA Steps and PROC Steps is not often clear. And the DATA STEP is often viewed as something a bit mysterious dataset, _eof_, proc step source 7
234 Tips and Techniques for Using I/O Subsystems in Version 6 of the SAS System under VMS. Darylene Hecht This paper discusses features of the external I/O subsystem in Release 6.06 of the SAS System under VMSTM, The filename statement, file statement, key, key=, infile statement source 7
235 Tips and Tricks on Rearranging and Manipulating Data and Datasets for Efficient Data Processing and Output Generation Donald D.M. Tong,
Thomas E. Schneider,
Pan-Yu Lai		 proc step, binary variable, proc transpose, proc contents, macro definition source 4
236 Transfer Function Model for Gloss Prediction of Coated Aluminum Using the ARIMA Procedure Mozammel H. Khan time series analysis, correlation, regression analysis, first difference, time series data, proc calis source 6
237 Transferring SAS Data Sets with the XCOPY Procedure Thomas A. Miron There are a number of reasons to transfer SAS datasets between an OS mainframe and a Pc. SAS provides PROC UPLOAD and DOWNLOAD to facilitate this transfer. While UPLOAD and DOWNLOAD are convenient and relatively straightforward they are not as efficient in terms of system resources as native file transfer programs such as IBM's IND$FILE. This efficiency difference can be significant when transferring large SAS datasets. In addition, UPLOAD and DOWNLOAD work via interactive SAS on the host system. Users with limited access to interactive SAS may not be able use UPLOAD and DOWNLOAD to transfer files. This paper describes an alternative method of transferring SAS datasets using PROC XCOPY on the host end, and PROC COPY and LlBNAME with the SASV5XPT option on the PC end. Actual data transfer is accomplished with the system native file transfer program proc step, dataset, proc upload source 4
238 Transporting Files in Release 6.06 of the SAS System Duane E. Ressler,
Holly Whittle		 This paper explains which tools you should use to create transport files to move SAse data sets and catalogs in Release 6.06 of the SAS System from one host environment to another. This paper also briefly addresses the topic of converting and transporting SAS data sets from Version 5 to Release 6.06. The examples in this paper illustrate movement between the MVS and the VMS~ operating sys tems, but the techniques discussed are applicable to other environ ments libname statement, proc step, catalog entry, dataset, proc copy, filename statement, catalog, select, proc cimport, proc cport source 9
239 Tuning Version 6 of the SAS System Under MVS Daniel J. Squillace This paper presents information on tuning the SAS System in both interactive and batch modes under MVS. Principal areas covered include memory and program management, I/O subsystem characteristics and performance, and performance aspects of some SAS System options. Architectural and implementation differences between Versions 5 and 6 which must be understood to effectively tune the system are also discussed options statement source 6
240 Unbiased Estimators of Shrinkage for Chi-Square Based Measures of Association Richard F. Haase proc freq, sample, correlation, proc cancorr, numeric functions, sample size, monte carlo, anova, multivariate analysis, degrees of freedom, response variable, binary variable, predictor variable, parameter, regression analysis source 6
241 Updating Your SAS Training Program for Release 6.06 of the SAS System John C. Boling Release 6.06 of the SAS System represents a new software design for the decade of the 1990's. A number of major design goals and software re-engineering in Release 6.06 necessitate changes to existing education offer-ings on the SAS System as well as suggest new offerings for SAS training programs dataset, proc step, proc sql source 3
242 Using a NeXT "Workstation as a Development Platform for Version 5 SAS Applications Joseph E. St. Sauver SAS Institute has yet to announce any firm plans to port the SAS System to NeXT workstations. Nonetheless, a NeXT workstation can serve as an excellent platform for developing VAXNMS· (or other mainframe) SAS System code for remote execution. The combination of a strong windowing environment, display PostScript support. a built-in athemet interlace and copious slorage eapacny bundled on lOP of more-or-Iess BSD 4.3 UNIX" make development of SAS System code on the NeXT for remote execution on another mainframe quite easy. The author's experience with use of a NeXT as a remote code development plaHorm for SAS and SAS/GRAPH" on a VAXNMS system is outlined, and some SAS·from·a·NeXT tips and techniques are described. While this discussion is couched in the context of working from a NeXT workstation, much of what is said is equally applicable to other networked windowing workstations source 6
243 Using a SAS/AF Application to Calculate the Power of a Statistical Test John F. Schneider,
Robert A. McLean,
William L. Sanders		 Estimation of sample size by considering power is a concept familiar to many researchers and statistical consultants, but is rarely used. Most avoid the tedious calculations. Recently, O'Brien (1986) and Sanders (1989) have demonstrated how the SASfD DATA STEP and PROe GLM can be used advantageously to make these estimates. This paper demonstrates that an AF application, which takes advantage of Screen Control language (SeL). simplifies this process and makes it more convenient for users to calculate power main effect, sample size, sample, power analysis, proc genmod, proc step, proc glm source 6
244 Using Relational Database Technology and the SAS System to Develop SAS' System Applications under the VMS "Operating System Jean Moorefield Mary works for the headquarters of a national advertising firm. She has been asked to collect data from each of the satellite offices to produce a firm-wide marketing analysis. Although each office shares computer resources, each has chosen a different database manage ment system (DBMS). Some have selected RdbIVMS7 while others use Oracle~ Mary herself prefers the SAS~ System. While Mary has access to all the data she needs, they are aU stored in incompatible formats and will require a lot of her time and resources to combine into one DBMS. If she could only find a way to bridge the communi cation gap and bring them all together, she could produce the reports she needs in minutes with the SAS System. With the release of Version 6 of the SAS System, Mary's problems are solved. The new release features Multiple Engine Architecture, designed to meet several goals including a transparent method of accessing data stored in the DBMS of other vendors dataset, where statement, proc sql, order by, proc step, variable, missing value, by group processing, update statement, proc append, set statement, proc sort, delete source 7
245 Using Release 6.06 of the SAS System for Financial Applications William S. Hopkins This paper will address setting up and maintaining a financial applications development environment using SASj ASSIS~ software as well as developing financial applications using this environment. A discussion on the use of SASjASSIST software to front-end base SAS. SASjFSp, SA~A~, SASjETS, SASj CALC" and SASjGRAPH software for providing so lutions to common financial applications needs is in cluded. Wherever possible, turn-key solutions avail able in SAS/ASSIST software are used thus keeping programming to a minimum dataset, proc step source 3
246 Using Release 6.06 of the SAS System for Marketing Information Systems Mike Senger,
Richard Roach		 Specific features of Release 6.06 of the SAS System are highlig,ted through a Marketing Information System (MKIS) case study. SAS/ASSIST"' software is used by sales and marketing profession als in developing applications that require virtually no knowledge of SAS language syntax:. SAS/ASSIST software activates several other Institute software products in this MKIS example logistic distribution, likelihood function, logistic regression, proc sql source 6
247 Using SAS Macros to Automate Data Set Creation and Analysis for Epidemiologic Studies on a PC William C. Murphy,
Howard M. Proskin,
Dennis H. Leverett		 This project was undertaken to automate the data-handling for a large epidemiologic dental caries study. In this study, many raw data sets wHh numerous variables are created from clinical exams In the field. Using the SAS· X statement wHh a rerouted DOS dir command, a list of the input files is generated and then fed into maCros wHh the SYMPUT routine. A master SAS data set is then created using a %00 loop. This data set has a considerable number of variables which characterize the dental state of the mouth. The PROC CONTENTS procedure Is utilized wHh an out option to generate a list of these variables that the SYMPUT routine uses to create macros for repetHive analysis. Again a %00 loop is used to pertorm repeated execution of a procedure for different variables. Using these methods, both data set creation and the analysis reqUire a minimum of user Intervention dataset, proc step source 2
248 Using SAS Software to Create WordPerfect Files Kathy R. Boussina put statement, audit trail source 3
249 Using SAS Software to Develop a Clinical Studies Status System H. Jean Lieverman clinical trial, dataset source 6
250 Using SAS/AF Program Screens to Create a Program Manager/ Document Producer for Data Editing Robin L. Wright,
Charles W. Kish Jr.,
Wayne M. Cannon,
Chris Bullard,
David Martinez		 The maintenance and documentation of data-editing programs have been greatly simplified by the creation of a menu-driven system called Program Manager/Document Producer (PMDP). This system, which uses SAS/AF soft ware, allows the Data Coordinators at A. H. Robins to eliminate repetitious programming tasks by utilizing PMDP screens to modify a set of editing programs. PMDP generates a document for archival purposes and a program directory for departmental use. Using PMDP has significantly reduced the programmi ng time per project and has eliminated a time-consuming process medical history, macro parameter source 6
251 Using SAS/AF Software in a Marketing Information System James L. Yates Most major companies have an enormous amount of useful information stored on their mainframe computers. Due to a lack of computer expertise on the part of most managers, however, this information can usually only be accessed via standard reports or predefined queries. The use of SASjAF software can change this situation in that it can provide the power of the SAS system to users who are unfamiliar with the SAS language. This paper describes the Sales Inquiry System that was written using SASjAF software. This system permits Marketing managers at Miles, Inc. to retrieve, summarize, and display detailed sales data by responding to the menu prompts. The user can retrieve and summarize on the basis of customer. date, product, product family, state, county, type of sale, etc., or any combination of the above. The information is then presented in either a table or a graph dataset, macro variable source 5
252 Using SAS/AF Software to Transport SAS Data Files from the VMS'"Environment to an IBM OS Environment Kathleen M. Freeman The Departnlent of Clinical Data Management at Merrell Dow Research Institute is using SAS software in a VMS environment to maintain SAS data files for clinical trials. In order to process large amounts of data into listings and tables for reports to the Food and Drug Administration, these files must be transported from the VAX to an IBM mainframe computer. This SAS/AF application was developed to automate the transfer of the SAS data files from the VAX to the IBM. Using information entered into AF screens designed for the project, this software system writes command files in VAX Digital Control Language (DCl) and IBM JCl to control1l1e transfer. These programs are automatically submitted to run in batch mode on the VAX and the IBM. A transport data set is created from SAS files and transferred to the IBM via an RJE con· nection. On the IBM, a two·step job runs, first creating a transport format data set using the IBM utility IEBGENER, and then new SAS data files using PROC XCOPY proc step, proc contents, proc copy, dataset, libname statement, adverse events, concomitant medications source 8
253 Using SAS/CPE Software for Analysis of VMS'"Computer Performance Data Tim Rowles This paper gives a brief overview of the capabilities of the 2.0 release of SAS/CPE" software running under the VMS- operating system. Each of the possible sources of raw data is discussed sepa~ rately and in comparison with one another. A discussion of the CPETOOL command and the CPE Menu System is included, along with examples of their use in data collection, management, and anal ysis. Examples of the use of SAS/CPE reports are included in a short discussion of the basics of VMS perionnance evaluation raw data, cpu time source 7
254 Using SAS/GRAPH Maps as a Nucleus For an Interactive Geographic Information System Jay Jacob Wind,
Jeff VanGorder		 We use SAS/GRAPH software to develop Geographic Information System (GIS) tools for federal and state environmental agencies. In this paper, we illustrate BAS/GRAPH maps, including 3-Ds, and show several special techniques. We demonstrate an interactive mapping ~ystem using SAS macros and SAS/GRAPH tools. We also discuss pitfalls you are bound to enoounter. This paper draws on our previous work, "Taking the Best Advantage of SAS/GRAPH Software's Geographic Information System Tools," SUGI 14, San Francisco, 1989, and "Using SAS/GRAPH Software to Integrate and Analyze Environmental Spatial Data," with R.W. Matheny, Best Contributed Paper, SUGI 13, Orlando, 1988 proc step, response variable source 6
255 Using SAS/GRAPH Software to Analyze Student Study Habits Bill Wallace This paper describes the steps taken to create unusual vertical bar charts for a set of data using an ANNOTATE= data set and PROC GANNO. Creating bar charts with the SAS/GRAPH PAOC GCHAAT is straight forward if the data, and the output desired, fall within PAOC GCHAAT'S capabilities. PAOC GCHART creates vertical bar charts by summarizing one or more vari ables from a data set. The bars onthe chart represent categories or discrete groups of one variable. The project described in this paper required one bar chart for each observation with the values of fourteen vari ables represented by bars. When PROC GCHART cannot create the chart requiredANNoTATE= data sets are good alternative proc step, proc gchart, bar chart, proc transpose source 6
256 Using SAS/GRAPH Software To Detect Medicaid Fraud Terry D. Allen bar chart, scatter plot source 7
257 Using SAS/OR Project Management to Plan and Manage a Major Electric Generation Facility Outage Wayne G. Maruska dataset, data wrangling source 6
258 Using the ADX Menu System for Taguchi Applications in Quality Improvement Randall D. Tobias regression analysis, main effect, anova, data science, degrees of freedom, posterior probability, standard deviation source 15
259 Using the Annotate= Data Set to Highlight the Graphical Quality of the G3D Procedure Kevin M. Hillstrom Annotate = data set can be used along with PROCG3D to enhance the visual quality of three dimensional graphics. Colors can be mapped onto the grid to represent countour levels, and annotate can place state boundaries on top of the contour to improve visual quality of the graphic proc step, if-then-else source 3
260 Using the DBLOAD Procedure to Create and Populate SYSTEM 2000' Data Management Software Databases David W. Pitts,
Kim D. Hiserote		 The architecture in Version 6 of the SAS'" System has opened up new ways to migrate data from SAS data sets to SYSTEM 2000" databases. It also provides a way to migrate data from other DBMS databases to SYSTEM 2000 software. This paper describes the process of taking existing SAS data sets, creating a SYSTEM 2000 database, and populating that database. Examples show how to create a database view, how to map data variables from a SAS data set to that view in order to add complete new entries in the data base. or to append records to existing entries. Input is not limited to SAS data sets. Any view supported by the ACCESS procedure in Version 6 of the SAS System can be used to populate a SYSTEM 2000 database dataset, proc step, sample, view, variable source 4
261 Using the FORMAT Procedure to Enhance the Subsetting IF Statement in Table Lookup Tom Krasner The technique of using PROC FORMAT with the PUT function for file lookup has been well documented in SUGI Proceedings. The technique uses the macro %MAKEFMT to create a PRQC FORMAT from a lookup file. A PUT function, which resolves the format on the basis of the value of the key variable, then constructs a non-key variable in the main file. A simple extension of this technique is to generate the format so that the value of the PUT function can be used as a boolean operator in a subsetting IF statement. The subsetting IF statement is then used to retain in the main file only those records which occur in the lookup file. The technique lends itself to a wide range of applications and often is the most efficient technique available. A further advantage is that once generic %MAKEFMT and %TESTIF macros are written using them is trivial. Thus at each installation a technical programmer could write the macros and provide novice progranuners with instructions on how to use them proc step, proc sort, proc format, cpu time, call symput, put statement, subsetting if, macro definition, macro variable source 6
262 Using the GSLIDE Procedure to Produce Data Collection Forms Mark E. Rubey,
Shirley H. Waters		 Many organizations collect data on specially designed forms for sales, orders, invoices, research, or other use~. One ~ay.to generate such forms is to wO:k wlth a ~rlntlng company, but th; s can be a tlme-consufln n9 and costly process. A more efficient way is tg produce the forms in-house using SAS/GRAPH GSLIDE procedure with the ANNOTATE feature. Rough drafts can be plotted quickly, adjustments can be made, and a final form can be ready to use in a short period. The advantage of this process is to have control of further changes and modifications sample, sas base source 5
263 Using the Interactive Features of the REG Procedure Rudolf J. Freund proc step, outlier source 8
264 Using the SAS System for Data Collection under the VMS'"Operating System Ed Blair This paper discusses functionality provided by the SAS" System and the VMS~ operating system that enables real-time or pseudo-real time data collection. An abstract data acquisition system model is developed and discussed. The concepts established in the model are applied to the development of a prototype set of user-written OAT A step functions designed to support the DataMyte~ Corpora tion 762 series data acquisition controllers. The resulting set of func tions represents a fully functional data collection tool kit. In the interest of setting the framework. for further development. the dis cussion remains abstract and general except for those areas where specific interface characteristics are required numeric variable, character variable, sysrc source 8
265 Using the SAS System on a Microvax for a Computer Assisted Interviewing System Veronica Fasullo,
Thomas Trumble		 sample, flat file, missing value, proc step source 4
266 Using the SAS System to Analyze Quality Control Data from Data-Collection Devices Julie A. Maddox,
Donna O. Fulenwider		 Since the introduction of SAS/QC" software into the SAse System, SAS Institute has received numerous inquiries concerning the use of data collection devices with the SAS System. Because automatic data collection allows for rapid gathering of large amounts of data with reduced potential for human error, these devices have become an essential element of quality control programs for many industrial and manufacturing sites. Combining the data.gathering power of these devices with the power and flexibility of the SAS System pro vides an ideal combination for manufacturing sites interested in quality improvement. To enable the SAS System to become a complete quality improvement tool, SAS Institute is testing Release 6.03 of the SAS System for Personal Computers with several devices from Mitutoyo Inc., DataMyte Corporation, and Ohaus Scale Corporation, world· wide leaders in the production of data·collection equipment quality control, dataset source 9
267 Using the SAS System to Develop Some Experimental CASE Tools Michael K. Burns,
Hoy W. Rogers		 CASE involves using computer technology to assist the software engineering process. In general it deals with all phases of software engineering from requirements analysis, structured software design, prototyping, code generation, test generation and execution, source management. and product maintenance. The thrust of this effort was to develop some tools to analyze existing software in it's source form and to build and maintain a knowledge base (called the repository) representing the internal structure of the subject software. This information details all named entities such as modules, entrypoints, global variables, macros, etc. 1\ also details the relationships of those parts. The information in the repository is available to the user for the purpose of fixing a bug, adding a feature, or redesigning the whole system character variable, proc step, dataset, numeric variable, macro facility source 10
268 Using the SAS System to Prepare Hospital Inpatient Claims Data for Case-Mix Analyses Daniel R. Bretheim outlier, proc step, sample, input statement, standard deviation, numeric functions, sample size, proc sort, proc freq source 8
269 Using the SAS/C Source Level Debugger's STORAGE Command Effectively Tim Hunter This paper explains how the STORAGE command in the SAS/C" source level debugger may be used to debug a storage overlay ABEND in a C program. The command format is described. Some fundamentals of heap and stack storage management by the SAS/C run time library are introduced. A general technique for debugging storage over1ays is explained data cleaning, char. source 4
270 Using the Semicolon: Put the Horse before the Cart and Other Tips for SAS Users Kimberly A. Foster User friendly programming should be the aim of all software development. The idea should apply not only to application programs for the end user but to the development of programming languages for use by application programmers. However, developers have given relatively little attention to the ergonomics of programming languages--how do the thought processes of the human mind affect user friendly programming? I start by describing how a minor change in the use of the semicolon provides ergonomic benefits to SAS programmers. I then follow with a potpourri of procedures I use, some of which are in direct conflict with recommendations made by authors at previous SUGI's and relate them to my own general philosophy of programming. We all have constraints imposed upon us and cannot always folloW every procedure in textbook fashion as we might desire. Part of programming is knowing the trade-offs involved. The Lowly Semicolon Ergonomics and computer programming dataset, input statement, macro call, macro variable source 5
271 Using the Stored Program Facility Joy Reel source 2
272 Using the Template Facility in the GREPLAY Procedure: Zoom, Rotate, Cut, and Paste Using the Panel Coordinates Arthur L. Carpenter The TEMPLATE facility in FROC GREPLAY is designed to display one or more graphs on a single screen or page. It is possible to position the individual graphs so as to produce a collage which creates the overall desired effect. At first glance many users find using the TEMPLATE facility difficult, however, a simple basic understanding of the relationship of the plot to the template coordinates makes the creation of interesting and unusual displays quite easy. Directed toward the new and intermediate user of PRoe GREPLAY, this paper explains in simple terms the steps required, from perception to plotting, for the creation of a tailored template. The primary examples used to illustrate the process include rotating within a template, zooming in on a selected portion of a graph and pasting the enlarged graph within a second template proc step, catalog entry source 5
273 Using the Version 6 SAS System with Your Operating System: Helpful Information Sources for Your Host Environment Dea Zullo,
Dee Stribling		 The presentation is a joint effort of tl1e Host Systems Research and Development Division and the Documentation Development Depart ment and describes the combined effort of host developers, the publications group, and the technical support group at SAS Institute to produce host-specific documentation and information resources that meet customer needs. This discussion shows the user how to maximize the utility and versatility of Version 6 of the SAS" System product by identifying where specific types of information are located with particular emphasis on host-specific documentation. The paper illustrates how the information interface and SAS soft ware reflect the uniqueness of individual operating systems, while preserving the continuity of the SAS System across host environ~ ments sample, options statement, sas base, sas log, libname statement, proc sql, macro facility source 7
274 Using Variance Components and Confidence Contours to Estimate Numbers of Primary and Secondary Sampling Units Andrew G. Stead In sampling situations. we wish to be within a certain distance of a population parameter with a certain confidence. It is often of interest to know, not only how many primary sampling units are needed, but also how large a subsample Is required. Many combinations of these sample sizes will satisfy the confidence statement To decide on what combination of these two sample sizes would be optimal, cost must also be considered. A sample, sample size, proc step source 5
275 What SAS/IML Software Has to Offer beyond the MATRIX Procedure Donna E. Woodward In Version 6 of the SAS'" System, the lML procedure replaces the MATRIX procedure. This paper briefly explores some of the enhancements of the SAS/IML" software product over PROe MATRIX and offers insight into their effective use. Induded in this dataset, proc step, proc iml, missing value, symbol table, local symbol table, input statement source 7
276 Working with Arrays Ronald P. Cody do loop, array, numeric variable, variable, character variable, data step, missing value source 7
277 Working with the SUMMARY Procedure: An Introduction John S. Boyden The PROC SUMMARY procedure allows the user to obtain statistical analyses on data obtained from a permanent, or working storage, SAS data set. This tutorial presents the basic concepts of using the procedure through examples. Using sample code followed by output produced through the procedure (printed using PROC PRINT), the reader may use the examples to see the cause and effect of most options of the procedure proc means, proc step, dataset, numeric variable, by group processing, output, class variable, variable, variable list source 6
278 Writing an 80386 C Compiler for the SAS System Glenn P. Musial This paper discusses the development of a C·compiler for the Intel 80386 microprocessor. A brief discussion of the evolution of the 80386 Is given, followed by a description of the 80386 register and instruction sets. From there, the C compiler's view of the 80386 is presented with emphasis on the features of the chip that make it attractive to a large-scale application such as the $AS system. In addition, comments on the future of the 80386, including operating system targets and processor evolution, are presented source 5
279 Writing Efficient C Code for Optimizers Oliver Bradley if-then-else, cpu time source 4
280 Writing Efficient SAS/DB2 Applications in Version 6 of the SAS System Charles A. Jacobs III The database engines introduced in Version 6 provide a way for SAS~ applications to easily interface with external database man agement systems such as D82: Because the interrace is a ~black box"to the application, it is difficult to know if the application is using the most efficient method to accomplish its task. The goal of this paper is to provide a glimpse behind the scenes into the DB2 inter face for Version 6, with an emphasis on choosing the best design strategies for different applications. This paper is intended for users who will be coding Version 6 applications using the SAS/ACCESS" interface to OB2 proc step, dataset, order by, where statement, if-then-else, subsetting if, select, proc sort, proc sql, cpu time, db2 access, work library, sorted data source 7

Back to Conf go 1991 go 1989

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.