HOME QUICK SEARCH AUTHORS KEYWORDS HELP BUGS

SUGI 1987 Papers - Index

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

Back to Conf
go 1988
go 1986
# Title
×		 Authors
×		 Abstract
×		 Keywords
×		 Extras Pages
1 "Beeperless "Programming-Effective Methods of Testing Using SAS Software Neil H. Howard,
James B. Pearson Jr.,
Linda Williams Pickle		 proc step, put statement, missing value, proc contents, proc print, sas log source 8
2 %QAFLIST-A Macro to Document SAS/AF Software Libraries R. David Sisemore SAS/AF is without a doubt one of the single most important improvements to the SAS product line because it provides the ability to tie numerous individual program modules into a COm plete "user-friendly" system. However, its biggest shortcoming is the inability to provide detailed documentation as to the contents of the libraries. This paper presents a SAS macro de veloped at General Dynamics/Convair Division to solve this problem macro put, proc step, proc cport, macro definition, %if-%then-%else source 6
3 A 'SAS UTM Projection for Geographical Information Systems Applications David J. Cowen,
James S. Roberts		 data wrangling, macro variable source 6
4 A Codebook Template for the Creation and Use of Permanent SAS Data Sets and Format Libraries (CMS Applications) John Podgurski proc step, proc format, dataset, raw data, input statement, label statement, format statement source 7
5 A Comparison of Table Look-up Techniques Craig K. Ray This paper presents table lookup procedures and shows how to use them efficiently with respect to CPU time. Table lookup can be defined as cross referencing a parameter file based on the value of a variable in the primary," (main) file. ~' When processing very large data sets, table lookup procedures must be carefully designed. * If designed poorly, these procedures can be a source of program bottlenecks. This paper presents table lookup tools and show how to employ them. As much as possible, the occasions to use one technique over another is presented in IIcookbookll fashion (f.e., IIfor lookup file of this size and for a main file of that size, Method A should be used"). This proc format, lookup, proc step, data step, dataset, call symput, char., cpu time, macro definition, point= option, format statement source 11
6 A Documentation Listing for SAS Programs William R. MacHose Document your SAS program wi th an atmotated program listing and word cross-reference. This poster displays,a SAS program that will document your other SAS ,programs. The atmotated listing highlights DATA and-PROC statements as well as all macro features. The cross-reference listing includes all program words not found in comnents. The documented program listing and cross proc step, macro definition source 3
7 A Flexible Mean-Standard Error Plot: Macro Program Will A. Sullivan,
Lianng Yuh		 A flexible mean-standard error plot macro program was developed using the HILOT and HILOe options in PROe GPLOT, Version 5. Using these options t the means of two treatment groups can be plotted with standard error bars so that the bars do not interfere wi th each other. This is true even If treatment means crOSB at any point In the plot. The macros are demonstrated with two biological examples. ·Furthermore, possible modifications for this macro program are also discussed. [Sec. 1] Introduction Mean-standard error plots are demanded for many types of aata ana~yses. In particular, the mean-standard error plot for two treatments at some specified time points or dose le~els. Most cit the eJl;:isting graphic packages' can produce mean plots with one-si<led or two-sided standard error ba~s; yet few have the ability to chose the direction of the bars so that one bar does not interfere with the other proc calis, autocall macro, macro variable, macro definition source 4
8 A Guide to Color Hard Copy Device Purchases Daniel Lee,
Irene Ortiz		 source 6
9 A Laboratory Notebook in SAS Software Caroline M. Bobik,
Arthur C. Singer		 The laboratory scientist and/or the research statistician is frequently confronted with the problem of conveniently documenting the description of experiments and presenting the proc step, r-squared, anova, proc genmod, proc print, main effect, data science source 8
10 A Macro for Estimating the Box-Cox Power Transformation Andrew G. Stead In their classic paper, Box and Cox (1964) demonstrated how a dependent variable could be transformed to satisfy simultaneously assump tions implicit in the analysis of linear models. For the class of analyses in which the response of interest is positive and where no transfor mation of independent variables is considered, a SASs macro is presented which makes the estimation of thi s power transformati on rel a tively easy. The use of the macro is illus trated with data taken from the original article of Box and Cox. Its application in investigat ing the effects of such a transformation on non-additivity is also demonstrated geomean, predictor variable source 5
11 A Manufacturing WIP Flow Model Created using SAS Software Barry Klipsch,
Mark S. Smith		 The inconsistencies in the flow of work through a manufacturing floor in a high volume, low similarity environment make WIP (Work in Process) projections difficult for short-term decision making. How much work will be in each of our manufacturing departments tomorrow, next week or next month? We needed to answer these types of questions to do effective workload planning. The IBM facility in Owego, New York is the major supplier of circuit card assemblies (pages) for IBM1s Federal Systems Division. We manufacture many unique CARDS of varying demand with each card moving through many different departments and build operations. When combined with the volume of individual pages processed, a large quantity of data is generated. We used extensively the merge capability of SAS Software to manipulate this large amount of data. SAS Software enabled us to extract and combine data and use the result to plot a WIP projection over a three month horizon using PROC GPLOT and PROC GCHART proc step, proc gchart source 3
12 A Menu-Driven CPE Reporting Application (The CPE Starter Set) R. Allan Russell bar chart, macro variable, proc step source 6
13 A Multi-Year Occupational Health Records System William M. Taylor Designed for governMent and industry to Monitor health risks associated with personal lifestyle or potential occupational hazards, and to provide eaployees with specific personal guidelines to iaprove their health, over 300 paraaetera are Monitored over a Multi-year periOd. Health profile data, including personal inforMation, aedical history, current syaptoMs, aedications, use of alcohol and tobacco, diet and exercise and ny other ite are coapleted by the eMployee on an annual baaia. Thi. data ia COMbined with a physician'. ex ** ination, blood laboratory work up, electrocardiogra., x-ray. vision and audio.etric teat. plus occupationally-related teats to establish * cOMprehensive databank for each person. The software, dependent upon tables and cOMputation foraula., relies extensively on the use of Macros. It is well structured to allow for easy update and revision management reporting, internal audit source 6
14 A New Approach to Creating Full-screen Menu-driven Systems for the SAS System under CMS Jeffrey L. Fowler While many approaches have been used in writing full-screen, menu-driven SAS" systems for CMS, none proved completely satisfactory for our purposes. Therefore, we wrote our own Display Manager: SHOWPAN. SHOWPAN is a SAS macro which displays menus and data entry panels from within SAS using the CMS full screen editor, XEDIT. It will work in any SAS 82 or version 5 system running under CMS Release 4. It was developed in an attempt to correct some of the problems encountered with other display managers. These problems are: The data set being edited must exist prior to displaying the screen. Code must be written for each screen to edit check fields and to validate responses. Response time is too slow. The screens take too much time to design. * Calculations cannot be done without returning to SAS. This paper describes how SHOWP AN addresses these and other problems and explains its internal workings dataset, macro definition, sample, error message, macro variable source 7
15 A Picture is Worth a Thousand Words and Numbers Michael W. Bornhoeft proc report, customer segmentation source 3
16 A Polygon Truncating Technique Provides Zoom for PROC GMAP Gregory C. Decker PRoe GMAP does not provide the establishing of a viewport in the world units of the map data. Hence it is not possible to extract a subarea of a map (zoom) by the bounds of some window which are independent of a maps poly gons. This paper describes a single DATA STEP that provides line clipping and polygon reconstitution. One can then graph only what one wants to see at the scale desired. Response variable features of PRoe GMAP, as all other features of PRoe GMAP, are retained. A working knowledge of PROe GMAP will be assumed if-then-else, proc gmap source 4
17 A Queuing Model for Analyzing Workloads of a Single-Threaded Process Jeanne J. Paynter,
Jerome P. Wood		 This paper discusses the effort and co m pletion of quantifying the performance impact of specific fourth generation languages on OST's CICS systems as well as informing management of the potential resource shortage to support these syste rns at reliable and consistent service levels of response tim e. Emphasis is placed on performance tuning and design improvements enhancing good performance as a re m edy to the im minent threat to providing these services. Rationale, methodology, and results are included. The primary tools are queueing theory, prohahilfty, numerical methods, the base SAS software, and the SAS/G RAP H system fullstimer, data cleaning, capacity planning source 7
18 A Rule-based Expert System Implementation for SAS Software H. Pat Artis if-then-else, variable source 4
19 A Sample Application for the New INFILE/FILE User Exit for IBM Mainframes Susan Marshall The new user exit for the INFILE and FILE statements allows users to read and write records from almost any type of data set within the SAS DATA STEP. The user can set up appropriate parameters and options, define SAS variables for such, things as input keys and feedback codes, print messages on the SAS log, and more. Technical Report P-156: INFILE/FlLE Statement User Exits fully documents this interface. This paper presents the features of this exit, discusses appropriate applications, and outlines a sample exit file statement, character variable, variable, numeric variable, put statement, infile statement, sample, execution phase source 5
20 A SAS Macro for Estimating Missing Values in Multivariate Data Gina M. Capalbo,
James S. Roberts		 missing value, sample, covariance, correlation, regression analysis source 3
21 A SAS Software Macro Application in Number Theory or SAS Software for Fun and Puzzles K.L. Kanthan SAS and SAS/MACRO language lends itself nicely in handlirg wide variety of problems. The intent of this paper is to show the flexibility of the SAS lal'lJuage in different areas such as m.unber theory, autanated JCL generation and solving puzzles. sane of the examples discuss areas where BAS is not typically used. Generations of large runbers is achieved without any exteooed arithmetic q,lerations and BAS is used to produce pericxiic decimals. Note that the SAS/MACRO language does not extend division facilities beyond integers. An example of SAS versatility is shown in autanatirrJ the JCL generation and to schedule job subnission which can be used for large systEms ~ reduce the operation involvement and manual prcxiuction run efforts. Third topic uses the recursive call feature of SAS/MACRO to solve a puzzle macro variable, macro facility, macro loop source 6
22 A SAS/AF Software Interface to Conversions Between FOCUS and SAS Software under CMS Dale C. Peters,
Charles F. Sprague		 This paper describes a utility to convert between FOCUS and SAS data bases using a SAS/AF interface and eMS EXEC2 programs. We first explain the rationale behind our facility and how we expect our clients at Pacific to use it most. Next, we describe some of the more interesting features of the interface, including a companion SAS/AF system that distributes updates of the utility and keeps track of its usage by clients. Lastly, we describe some of the enhancements we expect to introduce in the near future. Rationale for the Interface We work in a division of Pacific Bell concerned with managing information and systems for the Marketing Department. Specifically, our district functions as an "information center." Our responsibilities include guiding clients to appropriate data bases; building small-scale software applications. and; lending subject matter expertise, for example, statistical interpretations to reports and analyses proc step, dataset source 6
23 A SAS/ETS Software Forecasting and Inventory Planning System David Brauer,
Patrick Cardamone		 proc esm, time series analysis, time series data, proc step, residual, proc forecast, feature selection source 6
24 A SAS/GRAPH Software DISSPLA Interface using the Linkable and Metagraphics Drivers David G. Ball A device driver has been developed that allows SAS/GRAPH users to output plots in ISSCO's metafile format. This interface allows increased portabllity for SAS graphics output, as well as the possibility of taking advantage of a site's existing local and ISSCO-related software. The program was written in FORTRAN and has two versions. The first uses SAS's Linkable driver, while the second uses its Metagraphics driver. The Linkable version provides for 15 colors and is applicable only to IBM processors. The Metagraphics version can handle up to 256 colors, has potentially more capabilities (such as more dense shading), and is applicable to VAX machines as well as IBM machines. Aspects of installing the Linkable version at an IBM OS site are discussed. The implementation of a seven-color Metagraphics version for the VAXNMS system is described. One of the important features of these drivers is the proper translation of page size during the creation of the metafile proc step, batch job source 6
25 A Spreadsheet Application using Multiple Input Windows in the SAS System for Personal Computers Warren Repole Jr. proc format, proc step source 5
26 A Systematic Response to Customer Complaints using PROC FSEDIT, PROC FSLETTER, and PROC GCHART Gwen Bush This paper describes the development of a system to circulate customer complaint information, establish a database with the data, and statistically analyze the results using SASfacilities available on a Corporate IBM Mainframe. Repeated requests for statistical analyses of customer data in filing cabinets precipitated development of the system. The paper describes data entry into the database using PROC FSEDIT, forms development under PROC FSLETTER, and presentation graphics from PROC GCHART proc step, proc gchart, proc freq source 6
27 A VAX'M-VMS ""Command Procedure for Submitting SAS Software Jobs in Batch Mode James Barbour,
John D. Smith		 A VAX/VMS command procedure Is presented which helps SAS users submit JobS more easily and more reliably. When the procedure Is called as SETSAS It remem bers the name of the fl Ie containing your SAS code from session to session, the queue to which the job Is to be submitted, and the detal Is of any tape mounts that the jOb requires. It sets up loglcals CODE, LOG and LIS which point to the PLEASE SELECT SYSTEM: vaxc Username: SMIT~JD Password: corresponding .SAS, .LOG and .LIS files. When cal led as BSAS, the command procedure sub mits a SAS batch job, uSing the same parameters that SETSAS does. When no parameters are given, both SETSAS and BSAS retrieve the parameters from a file that Is written by SETSAS. II. Sample SessIons Two sample sessions are shown below to demonstrate the use of SETSAS and BSAS and the resulting messages. What the user enters Is In dicated by Italics batch job, sample source 5
28 Accessing SAS Software Sample Program Libraries Using CMS REXX Michele M. Burlew In order to provide easy online access to the SAS software sample programs distributed with SAS Institute's products, a system, EXAMPLES, was written in CMS REXX. EXAMPLES provides a means of reviewing an annotated list of the sample programs available, c looking at a specific sample program, copying a specific sample program, and printing a specific sample program sample, data step source 2
29 An Application That Interfaces PC and Host Resources Fritz Lehman dataset, proc download, proc step, missing value source 4
30 An End-User System for Reporting using SAS/AF Software and SAS/FSP Software as Code Generators Nora Lynne Yamasaki The end-user system creates and submits an ad hoc batch job coded in SAS software and the necessary operating system (JeL) statements. Run-time parameters are input by the user, who never sees any code, but instead works with menus with PF key selections, input screens, and when needed, help screens. The system, which was created using SAS/AF and SAS/FSP , was wr i tten for an OS/MVS mainframe using T50 batch job, data validation rules source 3
31 An Enhanced SAS/AF Software Consistency Error Checking Routine J. David Burks macro variable, %if-%then-%else, do loop, missing value, macro quoting function, macro definition source 3
32 An Example of an Integrated Programming and Data Analysis System Eric M. Klusman CERES, an integrated SAS programming and data analysis system, allows researchers to analyze pro duction databases in a large number of ways without needing to write the SAS code themselves. Users can choose from a variety of panels; each panel generates SAS code based on information from the user. Users can also write SAS programs directly on a SAS pro gramming panel. CERES builds a structured Browse Index of all submitted SAS code and the resulting SAS log and output. With the Browse Index users can re view or print SAS input, log, and output, or recapture . for resubmission any previously submitted SAS code. CERES was built by enduser programmers using IBM's ISPF dialog management facility and Tangram Systems Corporation's IIF, an interface between ISPF and SAS. Disclaitner. The opinions and findings discussed below are those of the author, and do not necessarily reflect a position of the Federal Reserve Bank of Chicago or the Federal Reserve System. Acknowledgements data science, sas programming source 8
33 An Integrated Audit System Based on the SAS System Kirk Paul Lafler,
David R. Mutziger		 Within the Quality Assurance Organ zation of Southern California Eqison company, a SAS-based integrated Audit System has been developed for use in scheduling and conducting audits. The Audit System provides comprehensive report, log, schedule and matrix capabilities for accessing current and historical audits conducted. This paper describes the major components of the system, features availabl~ to management and engineering personnel, and its evolution from a small prototype to an integrated tr~cking and reporting system quality assurance, macro facility source 4
34 An Interactive Method to Invoke SAS Software Report Programs Jon L. Bond This paper describes how a customer can enter parameters via the Interactive System Productivity Facility (lSPF) under the Multiple Virtual Systems I Extended Architecture (MVSjXA) operating system, causing a Statistical Analysis System (SAS) job to be submitted to create SAS reports. ISPF routines allow a menu-driven interface to receive parameters, which are eventually passed on to the modified SAS programs. In this application, previously written SAS programs are modified to include SAS macro-language statements. This paper focuses on the general methodology needed to use this technique macro definition, macro variable source 3
35 An Interface for EPIC and SAS Software James Lichtenstein This paper will describe an interface between SAS/GRAPH and EPIC Version 3.1 which will allow users to specify: GOPTIONS DEVICE = EPIC; in their SAS jobs and get graphic output on their 9700 laser printers. First some basic information about the EPIC software: EPIC runs in two phases. Phase 1 implements Calcomp type calls which may be used in a high level language (Fortran. PUI etc ... ) to define a picture and outputs a Vector File as a result. There is control information at the END of this file needed by Phase 2. Phase 2, using the control information and vectors created in Phase 1, creates a raster image in 9700 native mode suitable for sending directly to the printer. The important point here is that the vector file is used by both phases and that Phase 2 uses information at the end of the file to tell it what to expect in the file from the beginning to that point phase ii, phase i, proc step source 11
36 An Introduction to SAS/DMI Software Kerrie Brannock,
Glenn H. Itano		 sysrc, variable, proc step, sample, macro variable source 5
37 An Overview of the SAS System Random Number Generators Bart Killam The SAS USER'S GUIDE: BASICS, Version 5 Edition, lists nine random number generators (not counting the UNIFORM and NORMAL, which are older implementations of generators for the uniform and normal distributions respectively). Of these, five are correspond to common continuous distributions in statistics. The importance of the RANUNI generator lies in the fact that all the other continuous generators use RANUNI and transform methods to generate their respective streams. The RANUNI generator is a member of a general class called "Multiplicative Congruential Generators with Modulus 231 _1". A description of the algorithm for these generators s given, along with a description of the transform method used to obtain the RAN NOR generator from the RANUNI. Finally, some common missuses of the random number generators in SAS programming is discussed source 7
38 Analyzing Your Information Needs Richard W. La Valley The synthesis of data into "useful" and "usable" information is different for each distinct level of management in a company. The usefulness of the appropriate information is not apparent unless the presentation of the data is different for each level of management. A successful analyst must develop a general framework for the synthesis and presentation of data, and this framework should take into account the different informational needs of a heirarchical management structure in a company. This paper will attempt to present some practical guidelines of what data should be presented to each level of management. In addition, the paper will attempt to give some practical ideas of how the to effectively present the information to each level of management management reporting, mis source 5
39 Application of Intervention Analysis to Power Plant Monitoring Data Linda E. Bireley estimate, normal distribution, time series analysis, sample, proc step, regression analysis source 9
40 Application Prototyping in a Decentralized Environment Donna Breslin source 2
41 Automating Documentation: Or How to Avoid the Really Dull Work and Still Know What You Are Doing Fred Gehm audit trail, proc step source 5
42 Backing Up SAS Software Files Martin Fischman proc copy, proc step source 1
43 Batch Environment SAS System as an Aid to Managerial Decision Making Virginia N. Adams sample, dataset source 2
44 Batch File for Sharing SAS Software for the Microcomputer on One PC Denny Mills sas base, proc step source 2
45 Beyond Default Colors: Introduction to Customizing Graphics Output Janet Lasher This paper serves as an introduction to customizing SAS/GRAPFI output. It covet;s the basic form of the most frequently used and flexible SAS/GRAPH statements which are used to manipulate the default graphic output with simple techniques readily achievable by the beginning SAS programmer. It is assumed that you already know how to create default graphic output, and you are now in the position of trying to alter these graphs. Introductign Not all of the options available with the statements described below have been covered. These options cap be found in the SAS/GRAPH manual. It is my intention to explain the general form and methods of these customizing statements, not to cover every option available. I hope that with this introduction, you will have the confidence and knowledge to open the SAS/GRAPH manual and try graphics with a minimum of stress regression analysis, confidence interval, title statement, by group processing, options statement source 4
46 Building a Clinical Data Review System Using SAS/AF Software James F. Sattler proc step, feature selection, dataset, data review, clinical data, proc freq, by group processing, variable, verbatim term source 7
47 Building Custom-Designed Reports for Financial Executives Edward B. Dailey,
Jude L. Naes Jr.		 o Menus were too long and confusing. Reports with obscure descriptions were being run by mistake. Excessive time was spent keeping up with report requests. o We were losing management IS support for the SAS System. In order to regain support and get management to appreciate the capabilities of the SAS System, 708 we developed a report writer to give executives a method to run thei r own custom-desi gned re ports at any time. Our objective was to keep the report writer consistent and easy-to-use. We wanted to: Provide maximum flexibility. o Minimize typing. o Keep screens clear, orderly and consis tent. o Keep instructions and options brief and to the paint. o Display the report parameters as defined. o Provide shortcuts for the experienced person. o Provide help options where needed. Avoi d separate llhow_to ll i nstructi ons that could become obsolete. The finished product provides a tremendous amount of flexibility and, at the same time, re quires little typing macro variable, macro facility source 6
48 CARS - A Synergy of SAS Software, COBOL, and Data Analyzer for Health Care Reporting Kelly Tate,
Jim Woods		 CARS -- A SYNERGY OF SAS SOFTWARE, COBOL, AND DATA ANALYZER data cleaning, raw data source 5
49 Combining SAS Software and SYSTEM 2000 Data Management Software Mike Senger SYSTEM 2000 Data Management Software excels in the produc tion data processing world, of entry, retrieval, integrated applica tions development, and data management. The- SAS System offers tremendous versatility in data extraction, analysis, and pre sentation. The complementary nature of these two SAS Institute product lines is exemplified through day-ta-day management activities of Garcia Truck Leasing Company officials. Some mar keting and accounting problems are addressed by company man- agers with the aid of these software products. ' Gerry Garcia started Garcia Truck Lessing Company in 1977. He specializes in medium-sized trucks (10 to 22 feet in length) that pull trailers. The home office is in New York. The business has grown into a nationwide enterprise with offices in five regions. The company grosses about $9.~ million annually and is working hard to enhance its return on a considerable.investment in trucks and service centers proc step, dataset, proc freq, sas base, sample, proc glm, proc tabulate source 6
50 Combining SAS/GRAPH Software Output with Text Using IBM's Document Composition Facility James W. Mathews Integrating graphics into publications was a manual task, usually in~ volving paste-up, until we were able to combine SAS/GRAPH~utput with our text documents. We now produce quality documents (man agement reports, newsletters, papers, and documentation) using IBM's Document Composition Facility (DCF) and the IBM 3800 model 3 laser printer. The key to integration lies in the use of Page Segments, which can easily be produced using SAS/GRAPH once you know how. This paper will present the techniques involved in a compre hensive document, which should' eliminate most of the work involved in gleaning the necessary infonnation from the multitude of manuals provided by IBM and SAS Institute sas base, sample source 3
51 Comparison of Data Processing Procedures for the Statistical Analysis of Visual Field Data Katherine Freeman numeric functions, proc step, input statement, proc means, proc sort, macro definition, macro loop source 4
52 Comparison of Four Univariate Methods for Forecasting Seasonal Data by the SAS System Jeffrey Jarrett time series analysis, residual, sample, regression analysis, anova source 5
53 Comparison of PROC SORTT, Syncsort, and Plsort Janice M.B. Buck proc step, proc sort source 5
54 Computer Center Chargeback Using SAS Software and MICS Jeanne Buse This paper is an overview of the process of billing customers or users of shared Data Processing facili ties for the resources they consume. It concen trates on usage based, flat-rate billing systems where customers' charges are calculated by multi plying the amount of resources they used by the annual billing rate for the resource. This paper starts by reviewing the major billing functions. Then it shifts to the specific task of gathering usage statif>t.iCB for mainframe resources using IBM's! SMF records as a prototypal case. It discusses the steps required to process SMF records and identifies the associated problems and solutions. It shows how the SAS system solves many of these problems by itself. Also it shows how the HIes Accounting Component of Morino Associates Inc. of Vienna, Virginia uses the SAS software to solve the remaining problems and to provide a full featured billing system cpu time, by group processing source 5
55 Confidence Limits for the Product Moment Correlation Coefficient Based upon Normalizing Transformations Herbert Hamilton sample, correlation, sample size, confidence interval, proc calis, regression analysis, %if-%then-%else, outlier source 6
56 Controlling the SAS System and Its Users An Information System Tutorial Thomas R. Hoffman raw data, data governance source 5
57 Conversion of UTM Coordinates to Geographic Coordinates for SAS/GRAPH Software PROC GMAP Displa David C. Chojnacky,
Ronald P. Tymcio		 The Universal Transverse Mercator (UTM) grid is used to reference all forest data collected in the Rocky Mountain States by the USDA Forest Service, Forest Survey P£oject. Display of these data with SASjGRAPH mapping procedures requires a conversion of UTM coordi nates to geographic coordinates ~latitude and longitude). In this paper a SAS DATA STEP is given for conversion of UTM to geographic coordi nates. Also given are examples of Forest Survey maps using the SASjGRAPH with converted UTM data sample, proc gmap, proc step source 6
58 Coordinating a Site License through a University Mainframe Computing Center Trina Hosmer source 2
59 Corporate and Area User Groups Charles E. Shipp The author has experience in starting and supporting user groups that include the Northrop SAS Users Group, the Southern California SAS Users Group, and the Northrop Macintosh Users Group. He is a member of the board of directors of the Los Angeles Macintosh Group (LAMG) and started the Special Interest Group connecting Corporate (Macintosh) User Groups. The difference between user groups and clubs is explained, and the different leadership styles are discussed. Techniques that build in risk but have potential for spectacular success are analyzed and contrasted with techniques for moderate success but a greater guarantee of continuing success. The paper discusses the benefits of user groups to the corporation, the individual user, the user's department, and to SAS Institute data wrangling, proc step source 4
60 Corporate Computing: A Working Blend of Mainframe, Mini, and Microcomputers E. Jeffrey Hutchinson The arrival, rapid growth and subsequent blossoming of the microcomputer as a professional workstation had dynamically Impacted corporate computing in ways never envisioned by the entrepreneurs who founded the microcomputer industry. The proper use - and potential abuse - of microcomputers in today's modern business and manufacturing environments is of vital concern to company executives as well as data processing managers. This presentation describes the working approach of large mainframe computing, specialized minicomputer applications and microcomputer power at a leading gas turbine engine manufacturer. To set the proper prospective, this paper will first present a overview of Avco Lycoming Textron and its Electronic Data Processing (EDP) Systems. This will be followed by a quick review of Avcols Mainframe, Mini, and Microcomputer systems. Finally the author will present his visions of microcomputer use and abuse, and a crystal ball look into the future of the industry artificial intelligence, data wrangling source 5
61 Creating a "User-Friendly "General Ledger Reporting System Jerry E. Clark ledger, phase i source 5
62 Creating a User-Friendly PROC FSCALC Application Patti Brideson Setting up a user~friendly PRoe FSCALC application requires knowledge of PROC FSCALC and the special tools available for developing such an application. First, the developers must think about how the software works. They are gOing to be pushing, or sending, commands to the command line using the program screen, and they must always be aware of which screen the com mand is being pushed to, as well as what effect the command has on that particular screen. Second, they must be aware of the special variables that are available within the FSCALC procedure that allow them to control the environment for the users. Finally, they need to be able to second-guess their users in order to program around any problems users may run into while using the application. Tools to use There are many tools available for your use if you are developing a user-friendly application for others. These tools are often used together to make the application easy to use proc step, comment statement, call symput, feature selection, macro variable source 6
63 Creating Selectively a Permanent SAS Software Format Library from a Humongous Dictionary Judith B. Gabor proc format, proc step, format statement, proc sort source 4
64 Data Entry Backup System: An Application of PROC FSEDIT in SAS/FSP Software Hugh G. McCabe This paper describes the application of PROC FSEDIT in SAS/FSP and Base SAS by an intermediate size health insurance company. The need for the application arises when our data processing company's production computer is down. During these periods our data entry operators are idle, claims become backed up and substantial costs are incurred. Wi th the acquisition of a new computer system in our Research and Development Department we decided to set up a claims entry backup system to be used when the production computer is down. PROC FESEDIT provided· the flexiblity to replicate screens with over 200 entry items and edit capacity to minimize data entry errOrs. The paper wi 11 examine the development and implementation of the project from screen creation to production of raw data files which are integrated back into production proc step, data validation rules, data step, dataset source 7
65 Decision Support Systems using SAS Software and ISPF (or How ISPF Tables Can Enhance Your SAS Software Life) Christine Hendrick,
David Howe,
Walter Mercier		 bar chart, proc step, dataset, variable, proc freq source 6
66 Determining End User Response Time Using NLDM and SAS Software Michael W. Yount Network Logical Data Manager (NLDM) is a software tool that collects end-user response time statistics from a Response Time Monitor (RTM). The RTM is a hardware feature that can be placed on 3274 control units and is standard on 3174 control units. Unlike some other network software tools, NLDM reported response times are measured, not calculated response times. The data collected by NLDM can be made available in several different ways. Of specific interest in this paper will be the ability of NLDM to write an SMF (type 39) record. A SAS program can be written to extract the data from the SMF record. The data can be used to identify problem areas in the network, to report response time 'information to management or users and to set end-user response time objectives sample, data cleaning source 6
67 Developing an Enhanced Facility for Installing SAS Software for the Microcomputer under MS DOS and PC DOS Kevin L. Kiser Installing the SAS system on a microcom puter is an eleven step process. The Microcomputer Resource Center (MRC) at the Uni versi ty of Wyoming (UW) has developed an enhanced facility elimi nating most of the work for the end installer. This paper presents the key considerations in installing SAS software under PC DOS, the assumptions governing development of UW's installa tion facility, SAS Institute's SASLOAD batch file, and a discussion of the influence of MS DOS upon installation sas base, sample source 6
68 Effect of Iteratively Reweighted Least Squares (IRLS) on Teaching A Course in Quantal/Categorical Response Rita C. Berg,
James E. Dunn		 proc nlin, proc step, binomial distribution source 6
69 Effective Analysis of Non-standard Clinical Data Joan Baron The trapezoidal algorithm will be discussed as an approximation to numerical integration in the context of the analysis of data from human clinical trials. Several nonstandard charac teristics of clinical data will be addressed. The algorithm will be implemented in SAse code using explicitly subscripted arrays auc, if-then-else, sample source 6
70 Efficiency and Performance Considerations for the SAS System under PC DOS Gordon W. Davis,
Scott C. Sweetland		 This paper is designed as a tutorial for programmers who desire to improve the perf:ormance of their PC SAS applications on currently popular microcomyuters. In addition, the issues we discuss wi 1 aid those who are facing the decision of upgrading their current machines or buying new ones. A variety OI microcomputer configurations and hardware enhancements will be explored that greatly decrease execution times of the SAS system. Tips will be provided for maintaining the microcomputers, with l'articular attention given to fixed disks. We w1.ll then look at the PC SAS system environment: how to customize the srstem for ease of use and how to manage m crocomputer resources such as memory and disk space. Finally, we will discuss some efficiency considerations for SAS statements used in the DATA STEP. Efficiency will be judged in terms of both execution time and economy of disk space by group processing, if-then-else, numeric variable, dataset, proc freq, length statement, decimal, select, libname statement, sas log, input statement source 9
71 Eight Key Questions on the Future Direction of Established SAS Hallett German data wrangling, sample source 5
72 Electronic Mail Using SAS/AF Software and PROC FSLETTER W.D. Hutchison,
Geoffrey Wheatley		 Electronic mail is beneficial to a broad range of users. An electronic mail system based on SAS/AF and PROC FSLETTER is easy to implement and has the added advantage of initiating new users to the SAS* environment. The purpose of this paper is to provide and describe the code used to implement an electronic mail system using SAS/AF and FRoe FSLETTER under MVS/TSO proc step, macro definition source 4
73 Engineering Applications of SAS/GRAPH Software Charles H. Markham Jr. Martin Marietta Corporation, under a contract to assist the FAA in making an upgrade to the nation's air traffic control system, has utilized the SAS system as an engineering analysis tool for the past two years. Four types of engineering applications for SAS/GRAPH software are described in this paper. The first section discusses SAS programs for constructing US maps with symbols, labels and lines added to show FAA facility locations, Area Control Facility (ACF) borders, and facility interconnectivity. Also included in this section is a program which uses templates 10 make complex maps. The second saction describes SAS/GRAPH capabilities available for making terrain charts. Programs are discussed for topographical maps, depicting surface features of an area, and cross-sectional maps of the terrain between two sites. The third section of the paper demol'lstrates a way to utilize SAS/GRAPH software for microwave site analysis proc step, proc gmap source 7
74 Enhancements to the Micro-to-Host Link Edmund Burnette,
Cheryl A. Garner,
Todd Lloyd		 The micro-ta-host link, part of the base SAS0System under PC DOS, can be used for remote SAS processing and for transferring files be- tween a microcomputer and a host computer. This paper focuses on recent enhancements to the SAS link facility, 'mdud'mg external file and catalog transfer, new sample scripts, additional script commands, and an asynchronous terminal emulator. Support for new communications CARDS and foreign keyboards is also mentioned briefly. Finally, a ref erences section is included that describes all available documentation on the link sample source 8
75 Enhancements to the SAS Text Editor and PROC FSLETTER on the PC Claire Cates,
Joyce DeHaan,
Brenda Erickson		 source 4
76 Establishing a SAS Consulting Service at a Large Research University Ronald Kalinoski,
John Podgurski		 proc step, data science source 6
77 Estimation of Variance Components in Mixed Factorial Models Including Model-based Diagnostics Ronald H. Bremer,
John W. Green,
Ronald R. Hocking		 covariance, sample, proc mixed, anova, fixed effects, numeric functions, proc freq, outlier source 6
78 Experiences with Teaching Experimental Design Using the SAS System for Personal Computers Robert A. McLean,
Richard D. Sanders		 1986 - 72 programs - Over 3,000 students Institute Graduates - 3000 in 5 years Eight public programs per year 40 students per program Custom programs Design of Experiments 1986 Two public programs 33 students 1987 Two public programs 24 students per program Custom programs This introduction relates to much of the planning that was done before any attention was given to course content, educational philosophy, and the necessary type of computer hardware and software. The remainder of the paper will relate to these experiences. Our concluding section will illustrate a portion of the material that we use in response surface methodology and that will indicate the need for good software. Course Outline The following outline for the design of experiments course is based on the assumption that the student has been through a course equivalent to the three week Institute for Productivity Through Quality course regression analysis, contour plot, proc step, problem solving, main effect source 8
79 Exploring the Mandelbrot Set Anthony Patalocchi,
Paul B. Schladenhauffen		 proc step, format statement source 6
80 Financial Statements from COMPUSTAT Quarterly Industrial Data Files Mary S. Hirsch The COMPUSTAT financial data base is distri buted by Standard and Poor's Compustat Services, Inc. The expanded annual industrial file from this data base can be assessed using the macro SPYR as described in the 1982 edition of SAS/ETS. Following a similar format, two macros were developed to read the COMPUSTAT quarterly industrial files and output the selected data items to SAS data sets. The macros default to building data sets of all variables for every quarter for every company on the file. Options allow for selection of data items, years, and companies. A third macro, FINSTATQ, is listed that allows for inclusion of all data items in the IOO-item file pertinent to a balance sheet state ment. The data is transposed and labe11ed and thus presented in a familiar format macro quoting function, macro conditional, macro definition source 4
81 Forming SAS Time Series Data Sets from the International Financial Statistics Data Tape of the International Monetary Fund Alex Anckonie III time series analysis, dataset, time series data, input statement, time source 6
82 Frame It! Frame Generator Macros Akos Felsovalyi,
Erzsebet Felsovalyi		 introduces a SAS macro that frame' around the graph by a motif along the edges of the The frame motif can be chosen from 101 predefined motifs available for the user! The user can also define his or her own favorite motif by setting up a simple annotate-like data set. Another macro enables the user to select sections of the frame and to move them to a new location. Utilizing the latter macro, nested mul tiple frames can be created, or smaller parts of the graph can be framed proc step, input statement source 7
83 Full-screen Menu-driven Sample Size Programs Using SAS/AF Software J.A. Bebar,
Phil Cirulli,
L. A. Johnson		 This paper describes the efforts and results of writing a ful1screen menudriven application using base SAS and SAS/AF software. The application involves making the calculation of sample sizes required in various situations such as estimation and hypothesis testing for both variables and attributes as easy as possible. The goals were minimizing the number of screens a user had to go through before obtaining the sample, sample size, proc power and sample size, standard deviation, hypothesis testing, binomial distribution source 5
84 Generating Kappa Statistics and Testing Useful Hypotheses with PROC CATMOD Robert A. Terry This paper demonstrates how PROC CAl MOD may be used in the construction of generalized kappa statistics. Furthermore, it is shown how relevant hypotheses concerning interrater agreement may be tested using PRDC CATMOD, within the conceptual framework of the GSK cohen's kappa, proc step, weighted kappa, proc freq source 5
85 Generating Sample Sizes for Hypergeometric Sampling. Plans Using SAS Software Melvin T. Alexander This paper presents a method for obtaining hypergeometric attribute sampling plans. This method. an ex~ tension to a single attribute sampling plan developed by Alexander (1985). is useful in developing specialized plans for isolated lots. Examples are pro~ vided sample, binomial distribution, poisson distribution, if-then-else, sample size source 5
86 Genius Report Writer System Sue Trueblood Rakes The Genius Report Writer system is one of the retrieval languages available for the SYSTEM 200() Data Base Management System. This paper gives in-depth description of the Genius system and an overview of the various retrieval languages. SYSTEM 2000 DBMS offers a number of access source 2
87 Getting the Most from PROC TABULATE Andrew A. Norton proc tabulate, class variable, proc step, in=, missing value, by group processing, proc freq source 10
88 Graphic Display of Data in Clinical Trials Walter W. Owen Jr. Committees which monitor the progress and success of clinical trials often rely on the graphic representat ion of treatment group data to assist" in identifying tendencies that merit closer examination. Those who prepare the graphics need to present a maximum amount of data in an uncluttered format as well as expand and adapt the plots for tracking data over a period of several years. Several facilities within the SAS base product and SAS/CRAPH allow flexibility in tracking ~nd reporting longitudinal data. The capabilities of the PUT statement and the PRINT, PLOT, and CHART procedures are established but they are limited in their usefulness for production graphics. SAS/CRAPH allows customized graphics, with optional supporting text, to be created with the PROCs GPLOT and GCHART and enhanced with ANNOTATE= data sets. Text may be placed anywhere on the page and data points may be joined as needed proc step, clinical trial, proc gplot, scatter plot, regression analysis source 6
89 Graphic Profiling of User Group Activity H. Pat Artis standard deviation, sample source 2
90 Graphical Displays of Nonhierarchical Cluster Analysis Solutions John M. Larus This paper details a SAS/GRAPH~ macro library- that aids in determining the number of natural clusters present in the data. Two macros are available to display the results of nonhierarchical cluster solutions: TWOD which generates two dimensional scatter plots; and THREED which generates three dimensional representations of the cluster space. The variables used on the axes are selected by the user and these axes are automatically scaled. Optionally, users may select to represent cluster membership or display the dispersions of all points within a cluster. The use of PROe VISUALS~, an experimental SASl dynamic hyperdimensional graphics procedure, as a further aid in interpreting cluster analysis solutions is also demonstrated centroid, cluster analysis, proc step, proc fastclus, raw data, macro definition source 7
91 Graphical Illustration of a Water Pollution Index Using SAS/GRAPH Software Michael J. Millard,
E. Barry Moser		 A means of predicting and graphically illus trating degraded water qU~ity in estuaries was developed using various SAS statistical procedures and SAS/GRAPH software. Expected water quality conditions, as represented by a series of discrete and identifiable "pollution profiles II , are illustrated with GCONTOUR and G3D. Geographical orientation is provided by configuring SASMAP data sets as ANNOTATE data sets and overlaying with the data-generated pollution isobars or response surface. Mean conditions for each discrete profile are i11ust,rated, along with standard errors. This allowed easy identification of potential "trouble zones" of consistently high pollution levels, plus areas with high variability which may require concentrated sampling efforts in the future proc calis, sample, response variable, proc step source 4
92 Graphical Methods for the Analysis of Survival Data Frank E. Harrell Jr.,
Kerry L. Lee,
Barbara G. Pollock		 The analysis of survival-time data often involves an examination of relationships that are complex and difficult to display. This is particularly true when there is more than one covariate (independent variable) or when one of the covariates is continuous. Some examples of displays that are frequently needed but for which programming is tedious are: I. Plots of Kaplan-Meier [I] 2-and 4-year survival probabilities vs. sex and deciles of age. 2. Plots of one or more survival curves (with followup time on the x-axis) stratified by quartiles of blood pressure and adjusted for 5 other covariates. 3. Plots of Cox [2] survival estimates vs. a cont i nuous range of blood pressure, drawn separately by sex group. OS SAS currently has the tools for generating such plots (e.g., PROCs PHGLM, RANK, GPLOT), but they may be cumbersome to use regression analysis, proc phreg, proc step, proc reg, survival analysis, calendar year, class variable, macro definition, logistic distribution source 9
93 Graphical Presentation of Product Pricing Vicki G. Morwitz,
Theodore F. Rose		 Price sensitivity Meter (PSM), developed by Peter H. van Westendorp provides algorithms to determine various pricing points for a given product. Van westendorp's PSM algori thms have been implemented in SAS. SAS GRAPH is used to display the results. PROe GPLOT with cubic splines is used to smooth and display the various price curves. ANNOTATE is used to labe 1 the price curves, and to highlight various theoretical pricing points. PSM Description The implementation of PSM involves collecting respondents perceptions of product/service prices. Each respondent must provide a response to each of the following four questions: A. Print the letter "An next to the price where it is so expensive that you wouldn't even consider it. B. print the letter "B" next the price where it is so cheap that you feel the quality couldn't be very good. C. Print the letter "C" next to the price where it is starting to become expensive .* cumulative frequency, proc step, proc freq source 3
94 High-Performance Graphics for Exploring Multivariate Data Harold W. Gugel,
C. Ming Wang		 Graphical techniques have long been known as pow erful aids for data analysis. Numerous papers can be found on computer generated static graphi cal methods for data representation. Due to advances in computing, recent research in this area has principally focused on highly interac tive and dynamic methods. This paper describes a high-performance color graphics facility for exploring mUltivariate data. It was developed as a procedure under the SAS statistical system using commercially available software. Although designed to be device independent, it is most effective with high-end graphics devices. Through simple movements of a graphics locator the analyst can rapidly examine many different subregions of high-dimensional data. This facil itates the discovery and interpretation of struc ture imbedded in complex data data science, scatter plot, regression analysis, sample source 8
95 How to Use SAS Software Effectively on Large Files Juliana M. Ma dataset, proc step, production run, length statement, input statement, proc freq, format statement, proc contents source 9
96 Improving Statistical Methodology with the SAS System in an Industrial Environment Robert V. Baxley Jr. sample, anova, residual analysis, statistical software, proc step source 9
97 Incorporating Management Judgement in Forecasting Joel Fingerman predictor variable, confidence interval, interest rate, proc step source 6
98 Information Systems in the Utility Industry Based on the SAS System: A Southern California Perspective Jacob C. Epperson,
Roger B. Glaser,
Clark M. Roberts,
James H. Sorenson		 quality assurance, macro facility, sas base, macro definition, audit trail, data cleaning source
99 Input of Questionnaire Data Using the PROC FSEDIT Screen Michael P. Hudson label statement, numeric variable source 6
100 Instructional Building Blocks of Computer-Based Training Jane Krupnick advanced sas, libname statement source 4
101 Integrated Information Systems Kerry G. Bemis The goal of this paper is: goal describe information system that is proc step, macro variable, proc print, macro facility source 8
102 Integrating Information System Technologies: Laser Printing, Model 204 DBMS, and the SAS System Judith H. Mopsik dataset, data wrangling source 6
103 Integrating SAS System Analysis and Reporting into an ISAM Production Environment on the VAX Warren Repole Jr. macro variable, proc step source 4
104 Interactive Learning Curve Calculator Using SAS Software R. David Sisemore How much will the 205th widget cost your company to build? What is the average cost of any given production unit? What will the pro jected cost of any future production lot cost? These questions and many others can be answered by applying the 'Theory of Learning Curves', sometimes known as 'Improvement Curves'. Basically. a learning curve ;s a line on a graph that represents two primary facts [1J the time required to do a job decreases each time that job is repeated, and [2J that the amount of de crease will be less with each successive unit. Learning curve theory began as far back as the early 1920's, with the first published work on the subject in the February 1936 'Journal of Aeronautical Sciences' by Thomas Wright. Engineers and Project Managers recognized that employees engaged in manufacturing and assembly functions could work more quickly as more units were produced macro variable, macro flow control, macro conditional, macro definition, macro loop source 10
105 Interfacing SYSTEM 2000 DBMS with Fourth Generation Languages Gary A. Egan artificial intelligence, call records source 9
106 Interpretation and Diagnostics of the Multinomial Logistic Regression Dan Steinberg Multinomial logistic regression (MNL) is a powerful tool for the analysis of-problems with a categorical dependent variable having more than two values. It has had applications in evaluation of clinical trials, market research, transportation research, and has been applied. in virtually every social science and manag~ment analysis context. Nonetheless, most potentIal users are unfamiliar with the principles underlying the model, are not aware that there are two radically distinct versions of the MNL model . ' cannot Interpret the typically confusing output correctly, and do not know how to assess a model for goodness of fit. This paper seeks to address this problem by laying out the basic theory underlying the model, shows how the coefficients should be interpreted, and derives some useful diagnostics response variable, likelihood function, logistic regression, proc step, regression analysis, sample, logistic distribution, proc calis, goodness of fit, predictor variable, degrees of freedom, conditional logistic regression, variable, maximum likelihood estimation source 6
107 Intricacies and Ramifications of PC SAS System Management and Distribution at Large Academic Institutions Victoria W. Dingler source 3
108 Introduction to SAS/DB2 and SAS/SQL-DS Software Emily P. Wallace proc step, dataset, proc sql, where statement, select, sysdate9, variable source 5
109 Introduction to Version 6 SAS/AF Software Brenda Erickson,
Linda Wharton		 SAS/ AF software for personal computers combines a powerful programming language with the ability to easily create customized menu driven applications for end users. This paper introduces the new concepts in Version 6 SAS/ AF software, including new options in the BUILD procedure statement, the panels that make up an entry, the program screen language, and conversion of applications developed with Version 5 SAS/AF software to Version 6 format dataset, proc step, proc print source 5
110 Jittering and Other Graphics Macros for Exploratory Data Analysis Lawrence H. Muhlbaier There are many techniques for exploring the structure of data that are awkward to use with existing SAS*~procedures, as pointed out so graphically by Paul Tukey at SUGI 11. This paper discusses'implementatton of a macro to make some of these methods of exploratory data analysis (EDA) more accessable to the SAS user. In particular, a SAS macro for JITTERing in one dimension, with or without grouping levels is provided. EDA macros and procedures available to users in the SAS procedures, contributed procedures, and SUGI proceedings are reviewed box plot, proc step, data visualization, macro definition source 4
111 Lean and Mean: A Guide to Self Discipline in Using the SAS System Peter Kretzman This basic tutorial discusses tips for improving the overall efficiency of your SAS code. Too often, talking about "efficiency" in coding raises unwelcome images of abstract nitpicking just to save a millisecond or two of CPU time. With the power of SAS software, however, users can sometimes fall into bad habits that have a tremendous negative impact on the amount of work they can get done. When effi ciency considerations are totally ignored in SAS programming, bugs tend to proliferate and resource requirements can mushroom. Users then encounter what they falsely see as "limi tations" in the SAS software, shrug their shoulders, and give up. The goal of this tutorial is to show users spe cific ways to focus on tightening up their pro grams, leading to shorter code, less debug ging, and less CPU time for their SAS jobs. Rather than delving into obscure efficiency "tricks," the tutorial presents generally appli cable- ways to develop ongoing self-discipline in coding proc step, sas programming, by group processing, proc sort, macro facility source 3
112 Low-Dimensional Representation of High-Dimensional Data for Multiple Populations with Unequal Covariance Matrices Virgil R. Marco,
Danny W. Turner,
Dean M. Young		 A random vector is assumed to belong to one of several multivariate normal distributions possibly having unequal covariance matrices. The goal is to find a low-dimensional hyperplane which preserves or nearly preserves the separa tion of the individual populations. We present a computationally simple method of .deriving a linear transformation for low-dimensional repre sentation and give conditions under which the Bayes classification rule is preserved in the low-dimensional space. Finally, we utilize SAS/GRAPHS to present several examples to demonstrate the graphical low-dimension represen tation method sample, covariance, dimensionality reduction, proc princomp source 6
113 Making Health Analysts Independent: A User-Friendly Management Information System J. Patrick Nicol,
Leslie L. Roos Jr.,
Andre Wajda		 mis, management reporting, record linkage source 5
114 Making SAS/GRAPH Software Work for You Roger L Chenoweth proc step, proc gchart, macro definition source 5
115 Making SAS/IMS-DL/l Software User Friendly Henrietta Cummings The SAS/IMS-DL/I@ software product is an interface that allows processing of DL/I data bases through use of SAS@ program ming slatements under IMSfVS DB. IMS/VS DB/DC. CICSI OSfVS. CICS/DOSfVS, and DL/I DOSfVS systems. This paper demonstrates how to make the interface user friendly using the SAS macro facility that is a part of the base SAS soft ware. Introductions to IM3-Dl/l terminology, SAS terminology. and the SAS macro facility are provided for clarity macro facility, macro definition, input statement, dataset, proc step, put statement, file statement, infile statement, macro call, macro variable, macro parameter, %macro parameters, variable, proc print, sample source 6
116 Making Version 5 SAS Software Work for You: The Design of a Dynamic Reporting System for All Users Katie A. Hubbell Several Version 5 features of the SAS system have been combined to demonstrate one way to create a generalized, easily maintainable user interface without sacrificing flexibility and response to user growth. A reporting system is used as an example. The steps taken in designing such a system are outlined, and the use of SAS/AF, SAS/FSP, the Display Manager System (OMS), SAS macros including command-style macros, and AUTOCALL are highlighted macro variable, variable list, dataset, proc step source 4
117 Man Hour Estimating System Laura Donley sas base, dataset source 2
118 Management Graphics in a Quality Assurance Environment Shirley J. McLelland proc step, proc format, quality assurance source 6
119 Management Information Systems (MIS) for Information Centers using SAS/FSP SAS/GRAPH and Base SAS Software T. R. White,
D. E. Fowler,
Michael P. Farrell,
T. J. Gross		 The Carbon Dioxide Information Analy'sis Center (CDIAC), sponsored by the Carbon Dioxide Research Division' of the U. S. Department of Energy I h'as developed a Management Information System (CDIAC/MIS) that records and tr,acks a large number of the data center's varied information r,equests. CDIAC/MIS comprises three procedures: a menu-driven, front-end program written as an IBM/TSO CLIST; full-screen data entry and edit options written in SAS/FSPI and SAS execution modules written in SAS/GRAPH, SAS/BASE, or both. The CDIAC/MIS contains two SAS data bases: one records information requests and the other SAS data base is a distribution list for CDIAC's newsletter and other C02-related publications. ,The data bases can be manipulated to produce statistics and graphics that illustrate the information requests made by CO2 researchers and, policy makers and the types of persons on the distribution list source 8
120 Matrix Generation Using SAS Software Marc-David Cohen source 10
121 Maximum Likelihood Estimation and Simulation for One and Two Stage Sampling of Normal Mixtures Using SAS Software Kirby Jackson,
Wesley Kemmerlin		 In the investigation of a disease often a single continuous variable is measured for each individual and on the basis of the value of this variable an individual is classified as belonging to the diseased or nondiseased group. Assuming that the variable is distributed normally for each group with a possible different mean and variance then the observed data for a sample is distributed as a mixture of normal distributions. In addition if, in a second stage, a subset of the overall sample is measured on a vari·able that allows determination of the individuals true classification then more accurate estimation of the parameters for each normal distribution is possible. In this paper a procedure is developed using maximum likelihood estimation to estimate the parameters of the mixed normal populations under a multistage sampling procedure and perform simulations of possible sampling results sample, likelihood function, maximum likelihood estimation, normal distribution, heteroscedasticity, estimate, homoscedasticity source 4
122 Measures of Multivariate Skewness and Kurtosis Ronald Kalinoski,
NamJun Kang		 sample, outlier, sample size, covariance, likelihood function source 6
123 Mechanization of the Telephone Allocation Plan - A Network and/or Transportation Problem Carole Therrien The purpose of the telephone ~llocation process, an important part of corporate planning processes, is to outline the facilities needed to accommodate Teleglobe's traffic efficiently fOr the next 15 years in order to optimize use of available facilities under prevailing traffic condi tions. This paper describes the develop ment of a general model for a network linking Canada and a foreign country, and applies the TNETFLOW procedure from SAS/OR when flow through arcs of a network have to obey some additional linear constraints. Wi th the introduction of new canadian and foreign gateways, as well as new submarine cables and satellites, the trend indicates that more and more countries will have complex network configurations. In parti cular, with the multiple-foreign gateways models we face some problems such as the necessity of sharing telecommunioation facilities (submarine cables, satellites) to satisfy specific demand from each foreign gateway network optimization, sas base source 6
124 Micro-to-Host Application Michael Roberson The proliferation of personal computers has made effective micro-ta-host links increasingly Important. The SAS System provides a set of tools for facilitating micro-ta-host applications. This paper shows how SAS/AF~software for PCs can be used to create a user-friendly interface to ~e SAS System's micro-to host features source 3
125 Micro-to-Host Link Performance and Setup David F. Kolb This paper is a [ollowup to the EnJw.ncements to the Micro to Host link paper and discusses link hardware performance and setup issues. Topics include some software design concepts, recommended procedures for starting the link, general suggestions and hints and some error recovery techniques. Also included are suggestions for working with customer support and usage of the link debug bits and the break window. Procedures for modifying the SAS System Micro to Host Link ASCII/EBCDIC character set translation and for defining an alternate 3270 keyboard are detailed. A few com ments regarding current link devices, protocol converter support and SAS/RTERM coax support are added. And fmally a look at some proposed link devices, some wish list items and future research direction. All comments regarding proposed or future device sup port are regarding experimental research only and do not imply future support by the Institute source 12
126 Monitoring Corporate Plan Implementation with the SAS System Kathleen D. Crook,
Mark F. Jackman		 proc freq, proc step source 3
127 Mr. TIB: A Test Item Bank for Physician Assistants Thomas P. Reardon data step, proc step source 3
128 MVS Security and the Use of SAS Software to Gather SMF Data Bruce H. Blank,
Mike Mirabile		 The purpose of this document is to identify and h~l~t the ilIlportsnt MVS system's security aspects. This peper does not serve as a ''cure all" for sewrity breames by any means, but rather as an in:formtive lock_ at the internal romponents of MVS and where security breaches ,can result. SJS Software is the tool that can be utilized to review many of the aspects relating to. assurance and review of the MVS system security oomponents. l!dditicnel1y, 00' (System Management l'llcility) provides the IIBjar inl"t to our SA'l Software rode fur assunnce and review purposes, since 00' collects and rerorda a wide variety of related infurIIBticn. '''!be protecticn of data against persros deliberate or accidental access by unauthorized peraons is rapidly becx:ming a IIBjar preb1em. ULtiIIBte1y, the security of data dependa en some ronblnation of lod<s, or access-crotrol measures, for W:tid1 certain users possess the keys. lib such ronbinatien is cx:mp1etely secure access control, data wrangling source 6
129 Network Performance and Resource Management David R. Daner ledger, capacity planning, extract source 7
130 New CBT Features in SAS/AF Software for Version 6 Philip James Busby source 5
131 New Developments in Psychometric and Market Research Procedures Douglas P. Kent,
Warren F. Kuhfeld,
Forrest W. Young		 SAS Institute is funding a software research and development project at the University of North Carolina at Chapel Hill whose purpose is to develop a package of Psychometric and Market Research procedures. To date, this project has developed three SAS' procedures, one stand alone program, and one SAS macro. Four addi tional procedures are under development. In this paper we present a brief overview of the entire project, and then discuss PROC CONJOINT and PROC TRANSREG, new procedures for trans for.ation regression which performs conjoint analysis, preference mapping, external unfold ing, 'nonmetrip regression, and a wide variety of additional analyses. 1 source 6
132 Nonparametric Density Estimation C. R. Longbotham Nonparametric density estimation is a recently applied statistical tool useful in exploratory data analysis and in graphical present~tion of data. The underlying density of the data is estimated without assumptions regarding the form of the true distribution except that it is continuous. We describe a macro that may be called to give a nonparametric estimate of a univariate density. An introduction to nonparametric density estimation is included sample, if-then-else source 3
133 Numeric Precision Considerations in SAS Software Richard D. Langston source 6
134 On PLEX and Custom SAS Procedures Paul Kent The SAS(£) System allows users to write programs in third generation languages (PL/I, C). These programs are interfaced to the SAS supervisor and operate using the same familiar SAS syntax. The SAS System to SYSTEM 2000 DBMS interface procedures are general purpose tools that serve it wide variety of uses. However, they may not be suitable for some situations. Recovery and error checking are impor tant in ma.ny update situations. vVe consider a. program tJUtt must update the employee da.ta base from informa.tion supplied a.~ a SAS data. set. The program itself is quite simple: it updates the em- o ployee record) based on the employee number. However, it does illustrate all the points thaJ. need addressing when writing a user-written procedure. All you need to lmow about writing user written pro cedures is documented in SAS Programmers Ouide for PL/1, Version 5 Edition. Outline The program used as an examp.1e in this paper was writ ten for the SUGI demonstra.tion system source 6
135 One Approach to SAS Training Geraldine M. Resheske proc step, by group processing source 5
136 Optimizing IBM PC Systems for SAS Software Paul L. Marsh numeric variable, frequency analysis source 4
137 OS SAS Installation: Performance and Tuning Considerations Eric C. Brinsfield When you install the SAS System under OS, you must set many system options and choose several installation options. This paper discusses the implications of these choices and suggests some tuning techniques that could be used at your site. The paper a!so addresses some common problems encountered by Installation representatives over the past year options statement, cpu time, sas base, dataset, proc step, autocall macro, macro facility source 6
138 Panel Discussion E. Jeffrey Hutchinson source 2
139 PC Hardware Considerations for SAS/GRAPH Software Graphics Donna Bravo source 2
140 Performance and Feature Enhancements in the SAS System under VMS (5.16) Tom Cole sample, proc step, work library, dataset, sas base, proc printto source 4
141 Performing the Friedman Test and Associated Multiple Comparison Test using PROC GLM David Ipe proc glm, post hoc test, proc step, proc genmod, friedman test, degrees of freedom source 3
142 Permanent PROC FORMATs Eileen A. McGivern Permanent PROC FORMATs is.a programming technique used for storing PROC FORMAT code for repetitive use in SAS* programs. This permanent storage is accomplished using a partitioned data" set (PDS) where each format name corresponds to a member name in the PDS. A 'format can then easily be referenced in a SAS job by including the code for the partitioned data set in the job control language (JeL) and using the c'orresponding PDS member name for the format name in the SAS code. This technique was employed for a user who had approximately 30 files with a combined total of about 4000 records that had to be continuously recoded as PROe FORMATs for use in SAS pro grams. The USer was experiencing a lack of data integrity, data redundancy, and program inefficien cy, due to repeatedly coding and updating the formats in multiple programs. Permanently storing the user's PROC FORMATs solved these problems in the following manner proc step, proc format, sas log, format statement source 6
143 Preparing and Presenting Management Reports using SAS/AF Software Darius S. Baer A major goal of end user reporting is to be "'user friendly". This can be accomplished only by requiring as little understanding about the mech anisms of the interactive application as possible. The end user should not need syntactic knowledge. A conceptual understanding of the application com bined with a recognition of variable names and values should be all that is needed to obtain full use of the application. The user should be asked only to choose from menu listings that in which he is interested. Furthermore, the end user should be supplied with menus that are dynamically modified according to who the end user is and what choices he has made in previous screens. If these funclions are included, the end user is both more likely to use the application and more able to obtain desired information quickly and with minimum preparation. These goals were met in an application titled "Man agement of Department Information" (MGMT) macro variable, proc step, feature selection, macro definition, call symput, sample source 6
144 PROC GREPLAY -- Developing a Presentation That Your Users Will Appreciate Nancy Palmer Rooth proc step, proc greplay source 7
145 PROC TRANSPOSE - A Powerful Tool for Data Restructuring Bernarr Pardo proc step, proc transpose, dataset source 3
146 Processing Large Data Sets into Customized Tables Tamara R. Fischell,
Elizabeth G. Hamilton		 proc step, proc freq, proc means, calendar year, proc printto, keep statement source 6
147 Processing SAS Software Code: Smart Filters Today, a Proposed PROC and a Smarter Display Manager Tomorrow John D. Smith SUbmitting SAS code to SAS for execution Is only one use for the code we write. This paper ex plores the Idea of SAS code as data about the work we are doing. If we explore these data- even with the simplest tools--we' I I learn useful things about our work. This idea is I I lustrated with examples of what it Is easy to do today, with proposals for what Is certainly possible with a little more effort and with a discussion of what would be desirable in the future. Think ing and talking about how we use SAS code as data today could help guide the development of more powerful tools for SAS users tomorrow. II. I nt roducO on Submitting SAS code to SAS for execution Is only one use for the code we write. We also read it and quote It to each other to describe what we have done, are now doing or wi I I do. Using the machine to scan SAS code can help locate a needed fl Ie, debug the code we're writing or give us a better understanding of code we've already written dataset, proc step source 5
148 Receipt Control and Assessment Monitoring for a Multisite Clinical Trial Norman A. Constantine,
Henk Pechler,
Michael S. Shing		 SAS* based systems to monitor receipt of assessment forms and compliance with assessment date windows for the Infant Health and Development Program are described. Upon receipt of an assessment form or missed-assessment report, the coordinating center data control staff log the appropriate information into either the forms-inventory or missed assessment inventory files. Three types of reports are produced monthly and disseminated to our eight study sites. The Window Monitoring List Report provides listings of form status for individual infants. The Window Monitoring Summary Report provides similar information tabulated by site and form. Finally, the Assessment Monitoring System Graphic Report shows plots of the percent of forms due which have been received, not received, and administered within the window, by week since the beginning of the study. Following the proc print, sample, temporary dataset, dataset source 5
149 Report Writing with PROC PRINT and PROC QPRINT James A. LeGoullon,
Debra Martinson		 °The PRINT and QPRINT procedures are used .to print;: data from a SAS dataset. The reports produced by these procedures can be as simple or complex as needed, de pending_ on the options used proc print, proc step, format statement, dataset, label statement, title statement, by group processing, proc sort source 8
150 Sampling Very Large Datasets from SAS Software: PROC RANDOM and PROC RANSTRAT William T. Douglass,
Gail Johnson		 sample, proc step, dataset source 4
151 Sampling with SAS Software: A Macro for Drawing PPS Samples Eleanor M. Dillon,
James E. Smith		 sample, sample size, set statement, systematic sampling, macro definition source 5
152 SAS Macros Designed to Create Camera-Ready, "Typed "Tables Directly from SAS Data Sets Tim J. Gabel There are often occasions in the reporting of analysis results when the following con straints apply: 1. A large number of tables are to be produced. 2. A specified format must be followed; i.e. tables are to appear in a fom not available through PROC PRINT, PRCC TABUIATE, or sare other convenient procedure. 3. Quality control neasures dictate thet the mmilers not be transcribed and typed manually into correctly formatted table shells. Normally, the use of PUT statements and a FILE statement would be the logical solution. PUT staterents offer a great desl of flexibility in the presentation of data. However I in the case of a large number of tables it can becane a te::iious, tine-consmning, and expensive process to set up job after job of PUT statements. SAS Macros can be used to speed things up greatly. This paper will present several techniques and suggestions to aid progr"""'""" in designing macros to produce carrera-ready tables macro variable, macro function, macro facility, proc step, macro definition, proc format, put statement, character functions, label statement source 6
153 SAS Prototype Applications Bringing Power to the End User Richard Roach proc step, sas base source 3
154 SAS Software Data Sets from NOMAD2 John M. Rinehart This paper discusses the use of SAS Software in' a research environment for the analysis of' data stored in NOMAD2 databases. The rationale for segregating data storage and data analysis Is presented, along with an outline of the process of transferring data from a NOMAD2 database to a SAS dataset. NOMAD2 procedural code Is provided which largely automates this process, making aqy permanent or created NOMAD2 database segment readily available as a SAS dataset. The research data peoressing environment RMC Environmental Services 1s a private environmental consulting firm providing study design, data collection, data processing, analytical and interpretive services for environmental research and mOnitoring programs. One of our major clients for the past two decades has been Philadelphia Electric Company (PECo) dataset, variable, input statement source 4
155 SAS Software Implementation of Rockart and Treacy's Data Cube: a Typical Executive Information System Paul OldenKamp proc step, dataset source 6
156 SAS Software Installation-Things Your Site Representative Never Told You John W. Davison Jr. sample, proc step source 6
157 SAS System Interfaces to Data Base Management Systems Emily P. Wallace dataset, proc step, where statement source 4
158 SAS System Statistical Procedures Applied to the Analysis of Water Quality Data Paul D. Mowery discriminant analysis, proc step, response variable, correlation, sample, proc candisc, proc discrim, proc glm, covariance, proc genmod source 10
159 SAS System to IDMS/R Interface Overview Nancy J. Wills This paper is designed to provide an overview of the current development being done on an interface between the SAS Sys tem and IDMS/R~ The paper is geared to experienced SAS users who understand the concept of a data base and are familiar with IDMS/R. IDMS/R is Cullinet Software, Inc.'s data base management sys tem (DBMS). It runs under as, DOS/VSE and eMS operating sys tems. IDMS/R has both network and relational capabilities. The data bases are controlled by IDO;r-the Integrated Data Dictionary. The interface between IDMS/R and the SAS System is being developed because of the interest expressed by both SAS and IDMS/R users. We have received numerous letters. phone calls, and verbal requests from customers. The interface was a very highly requested item on our SASware Ballof.ID Let's take a look first at the network capabilities of IOMS/R. To begin with, let's define a set. A set is a relationship established between two or more record types dataset, format statement, where statement source 2
160 SAS Training in an Information Center's Diverse Environment Sandra Robinson,
Sallie R. Teaf		 Although information centers seem to be flourishing and continue to grow in number in business and academic settings across the coun try, clients and information center staff identify training as a weak point in most information centers. At The University of New mexico we have made training a focal point for improvement within the Information Center by designing a multilevel, integrated SAS training program to meet the diverse needs of our great variety of academic and administrative users, This paper demonstrates how we are successfully achieving our two main objectives: quality training and self-sufficient clients sample, dataset source 3
161 SAS/FSP Software, SAS/DMI Software, and SAS/AF Software as Tools for Work Stations David T. Bean dataset, proc fslist source 7
162 SAS/GRAPH Annotate Application Sheila Evans proc step, %if-%then-%else, macro definition source 9
163 SAS/GRAPH Software: Data Analysis, Presentation Graphics or Pretty Pictures? Paul L. Marsh bar chart, vertical bar chart source 6
164 SAS/SHARE Software Overview and Features Bill Brideson proc step, sas library, dataset, sas programmer, sas log, by group processing source 6
165 Screening for Potential Confounders in an Epidemiologic Analysis Linda Williams Pickle odds ratio, data analyst source 4
166 Seasonal Adjustment Operations of Canadian Socio-Economic Time Series Estela Bee Dagum,
Misako Grafton		 This paper purports to discuss the opera tions involved in the production and dissemina tion of seasonally adjusted data, with special time series analysis, seasonal adjustment, proc step, moving average, seasonality source 5
167 Simulating 2-D Gel Electrophoretograms using Techniques from Version 5 SAS Software Michael S. Lajiness It is theorized that a comparison between the protein composition of normal versus cancerous tissue may provide a clue to the underlying cause of the cancer. 2-Dimensional (2-D) Gel Electrophoresis provides a method of separating the proteins for each tissue into a 2-D map. called an electrophoretogram. Examination of the map for a normal tissue overlayed onto that of a cancerous tissue is critical to this research effort. A commercially available microcomputer-based system can analyze 2-D gel electrophoretograms, scanning and quantifying each protein in the map. Unfortunately, some aspects of this sytem are unsuitable. However, one can produce an output file containing descriptors of all the spots on a map and then process it with Version 5 SAS software tools to provide the desirable functions. This paper will describe the use of SAS/ AF and SAS/GRAPH in this application. SAS/ AF is used to provide simple access to the available options proc step, sample, proc greplay source 8
168 Software Mirrors in a Mainframe/Microcomputer Environment Chapman P. Gleason data step, macro definition, macro facility, sas programmer, proc step source
169 Some Applications of PROC MATPAR: Match Pair Partial Correlation Procedure Chun Chen,
Robert E. Johnson		 correlation, partial correlation, proc corr, odds ratio, proc step, proc calis, sample source 6
170 Speech Recognition, the PC, and SAS Software Linda Fineman Over the past four decades, researchers have often thought that the solution to the problem of machine recognition of speech is "just around the corner." Realization of the dream has been a long t:me coming, but today speech recognition is available for limited applications on both large and small computers. At SAS Institute, developers are building a speech interface to the SAS System for personal computers using the IBM Voice Communications Adapter source 4
171 Speller: A Mainframe Spelling Checker Bruce Guthrie data step, txt source 6
172 Strategies for Development of a Corporate Graphics Center Suzanne S. Gordon,
Jerry J. Hosking		 proc step, sample source 7
173 Strategies for Repeated Measures Analysis of Variance Phil Spector repeated measures, anova, correlation, repeated measures anova, sample, proc genmod source 4
174 Supporting End-User Computing through Full-Featured SAS Files Robert Langill Giving people access to the inFormation they need is the basic mission of data processing departmeots~ To perform this mission, DP professionals often employ interface tools desIgned to shield people "ho need the information from the camp 1 ex Mays .... e store it. We use menu driven systems, code gener ators, query systems and other devices. While it may lack some of the flash of n~er tools, base SASR soft"sre provides people Mith a Kay to obtain and use the information they require. And, a SAS file that provides a high level of self-documentation and simple access to its information is an effective user interface tool. This paper is a study on a set of ful1- featured SAS files. These files incorporate general design and special SAS £eatures that provide access to in£ormation "Ith simpll£ied coding and improved i n £ormat i on output. With these £ lies, peep 1 e use the speed and £lexibility o£ their own SAS jobs to produce the reports and ana 1 yses they need missing value, proc step, proc contents, date variable, conditional do loop, numeric variable source 6
175 Supporting SAS Software in an Academic Environment Leanne Whiteside This paper is intended to share the problems and solutions of educating computer users of the SAS. system in an academic environment. Briefly the problems were the obvious of how to educate students, faculty and staff with various levels of SAS knowledge. The range of instruction needed was wide, from what is the SAS system and why should it be used, to how to deal with large data sets on tape, to the specific problems of using the SAS system on the DEC (Digital Equipment Corporation) VAX. computer. Our problems were further complicated since we were not only new SAS users but also had new computer equipment and new operating system (VMS). The solutions involved the development of short seminars and workshops, various information handouts, DCL commarld files, as well as a staff person available for consultation with research and teaching faculty sas consultant, batch job source 5
176 SYSTEM 2000 Data Management Services and the SAS Macro Language Mike Killam macro facility source
177 TABLETS-A Table Generator Bernard Costa proc step, proc means source 4
178 Teaching Technical Subjects to Adult Learners Linda J. Stewart problem solving, quality of experience source 3
179 Techniques for Using SAS/AF Software More Efficiently Annette T. Harris System resources and software are just a few of the factors that impact decisions regarding development of new applications. You should take a closer look at the specifics of SASjAF soft~ ware in order to anticipate and plan for both needs and problems that may arise in a new system. Hardware Considerations You should consider every detail, even the hardware that will be used. How much does SAS/AF software know about the termi nals that are used? The configuration of the screen is stored with the screen, meaning that a screen created 9" a 30 X 132 configuration could not currently be displayed on a 24 x 80 config uration. Also, if you want to take advantage of colors and other highlighting attributes, the terminal and customization of your controller are important. If your terminal is capQble of using the extended highlighting attributes but is not using them you may want to check Item Number 160 in the customi?:ation of your control unit proc step, macro variable, put statement, data cleaning source 4
180 The Analysis of Mapped Spatial Point Patterns E. Barry Moser A SAS macro for the ana lys ; 5 of mapped spatial point patterns is described. The nearest nei ghbor and nearest event di stance tests are implemented along with second-order tests described by Ripley (1977). These tests permit the classification of patterns into aggregated, regular, or random point patterns. Monte Carlo tests permit further testing for randomness as well as testing against other alternatives. The SAS/GRAPH procedures are used to di sp 1 ay these functions to aid with their interpretation. A Poisson cluster process is fit to redwood seedling data using PROC NLIN monte carlo, poisson distribution, proc step, probability distribution, model training, macro definition source 5
181 The Analysis of Survival Data with Concomitant Information using Parametric Models: PROC HAZARD and PROC HAZPRED Eugene H. Blackstone,
David C. Naftel		 PROC HAZARD and PROC HAZPRED are two user contributed procedures that are used for the parametric analysis of survival data including the incorporation of concomitant Information. The parametric model is a flexible model that consists of up to three phases that are additive in the hazard domain. Each phase is scaled by a log-1 inear function of the concomitant information. PROC HAZARD builds the model and estimates the shaping and regression parameters. It contains several stepwise options and several different algorithms for optimization. PROC HAZPRED computes predictions for the survivorship and hazard functions along with their confidence limits. Input for HAZPRED includes model specification, parameter estimates and covariances, and a set of observations consisting of values for time and for the concomitant variables in the final model proc step, covariance, likelihood function, estimate, proc calis source 6
182 The Bayesian Estimation of Linear Models: An Application of the Lindley-Smith Hyperparameter Model Dennis J. Aigner,
Joseph G. Hirschberg		 covariance, estimate, sample, regression analysis, proc step, proc reg source 6
183 The Busman's Holiday: Automating the Systems Development Process Roger E. Booker,
Julie Keenan		 data cleaning, dataset source 3
184 The Design and Programming of Large Systems using SAS Software John Kolman proc step, macro variable source 3
185 The Development of CHESMAP-A Program for Mapping Environmental Data Dan Jacobs,
David G. Swartz		 The development of a computerized approach to display environmental data in maps can provide an important tool to those interested in the analysis and representation of environmental data. Once the domain of cartographers or skilled graphics design personnel, customized maps are now available to anyone with access to the graphic capabilities provided by the SAS· System. This paper is a howwto guide for those interested' in producing specialized mapping programs. The example presented is a mapping system developed to represent environmental -data on the Chesapeake Bay. CHBSMAP was written using the macro facility features of SAS and the graphics procedures of SAS/GRAPH*. The user is only required to create data files of the envi ronmental data and select the appropriate CHESMAP options. CHESMAP will generate the SAS code necesary to depic't the data points, con tours or response surfaces in two- or three dimensional multicolor maps macro definition, macro facility, dataset source 5
186 The Evolution of a Plot Chris Potter time series analysis, line chart source 5
187 The Generation of User Defined Reports Using SAS Macros Andrew Johnstone In the past, reports have been generated by mapping on a piece of squared paper exactly how the report is to be laid out. The report is then written using WRITE statements or their equivalents (such as PUT in the SAS system) and specifying line by line the infor· mation to be printed, and the columns in which to print them. A user (or progranuner) could not .actually see the report layout properly until the program was at the testing stage. Conversely, panel generation routines have_ advanced to the point where now it is possible to 'paint' a panel so that the panel is generated as the user would see it. . This paper discusses the use of SAS macros to eliminate the re port coding step, and the design of a report layout using the 'painting' concept. The views expressed in this paper ~ entirely those of the author and in no way should be taken to, be the official views of the Re serve Bank of New Zealand, the author's employer dataset, macro variable, macro definition, proc step, proc print source 4
188 The Information Center: The Growing Pains and the Adolescent Years Henry B. Edwards source 3
189 The Information Centre at AEtna Life in Sydney Australia Linda Renneberg,
Peter Sweeney		 sas base, ledger source 4
190 The Internals of Macro Processing Lynn Hannah Patrick Once the basic concepts have been mastered, understanding the internals is the next logical step to enhancing your knowledge of the SAS Macro Facility. This tutorial addresses the internals by identifying the components of the macro processor, illustrating their relationships with each other and their interactions with the SAS supervisor, and examining how a macro program is processed. Major components include the input stack, wordscanner, word queue, macro compiler, macro executer, symbolic substituter and open code handler. Special topics in macro proce.ssing include the performing of symbolic substitution, resolution, evaluation, quoting and unquoting, and autocall processing &variable source 6
191 The SAS Supervisor Donald J. Henderson,
Merry G. Rabb		 dataset, set statement, keep statement, merge statement, by group processing, output, execution phase, do loop, drop statement, length statement, proc step, sas programmer source 10
192 The SAS System for Personal Computers and Electronic Data Recorders at a Field Location Russell T. Graham,
Jonalea R. Tonn		 The SAS System for personal computers (PCs) allows researchers to use SAS software at locations where mainframe access is difficult or impossible. In combination with electronic data recorders, PC SAS software provides an efficient system for collecting, editing, and analyzing study data. Data can be collected on electronic recorders and transferred by asynchronous communications software to a PC for editing and analysis. By configuring the communications port on the PC appropriately, data can be read directly into the Display Manager of PC SAS software. Attempts to read observations into the DATA STEP from the recorders failed, but if available, this capability would make the SAS System for personal computers even more useful in research data management and analysis raw data, _eof_ source 3
193 The SAS System under CMS: Tricks of the Trade Michael J. Deeney As powerful as the SAS System is, sometimes it I S easier to use the tools available within CMS to assist userS in the development and debugging of their applications. This paper describes several eMS e~t'ecs and XED~T macros which are used to"' invoke SAS/FSP and. 'the SAS help facility from OMS, as well as from XEDIT and FlLELIST. Their use in the ITG'consulting environment is emphasized proc step, dataset source 4
194 The Traveling Salesman Problem and the SAS System Akos Felsovalyi,
Erzsebet Felsovalyi		 The paper describes the SAS tools to solve the Traveling Salesman Problem (TSP) numerically and to present the solution in graphical form. The paper consists of two main sections: 1. Defines the problem and describes the possible methods of the solution. In troduces three SAS macros that yield the numerical solution of TSP. The solution is either exact or approxi mate depending upon the number of cities. Illustrates an interesting programming technique using the DATA STEP. 2. Presents some SAS graphic macros that generate the graphical solution of TSP. The graphic macros introduced here are more general. They draw the tour of the specified cities on the US map along with the names of the cities. The tour can be polygon-type or "hub & spoke"- type tour macro definition, output delivery system source 6
195 The Wild Side of. the Annotate Facility Chip Kelly proc step, missing value source 8
196 Time Series Intervention Analysis Using SAS Software Terry J. Woodfield predictor variable, time series analysis, proc step, response variable, proc arima, regression analysis, missing value source 9
197 Tools for Accessing SYSTEM 2000 DBMS Data Using SAS Software Henrietta Cummings,
Nancy J. Wills		 dataset, proc step, variable, where statement, sas output source 4
198 Training Manufacturing Management Using the SAS System Linda T. Jolley output delivery system, proc step source 6
199 Training Support for the SAS System for Personal Computers Janine J. Huff data wrangling, dataset source 3
200 Transfer Functions using PROC ARIMA John C. Brocklebank,
David A. Dickey		 correlation, proc arima, proc step, proc calis, regression analysis source 7
201 Treatment of Lagged Dependent Regressors and Autocorrelated Errors in SAS/ETS Software Thomas B. Fomby,
Joseph G. Hirschberg		 covariance, sample, regression analysis, proc step, proc autoreg, likelihood function, maximum likelihood estimation, response variable, predictor variable, proc reg source 6
202 Tuning the SAS System under CMS Daniel J. Squillace proc step, dataset source 4
203 Understanding the SAS Macro Facility Martin J. Rosenberg This tutorial introduces the SAS macro facility and describes what macros are, why they are necessary, common applications, and how SAS processes them. No prior experience with macros is necessary, but an understanding of the DATA and PROC steps is assumed. What Are Macros? Macros are a facility within SAS for writing flexible code. To illustrate what flexible code means let's look at a program written in a traditional programming language. PROC FREQ is a SAS procedure for constructing and analyzing contingency tables and is itself a program written in PL/I in the mainframe and minicom puter versions of SAS or in C in the PC ver sion. let's examine the two examples below. Example 1: Two-way table macro variable, macro facility, proc step, proc freq, macro definition, %if-%then-%else source 6
204 Use of Base SAS Software as a Teaching Language for an Introduction to Information Systems in a University Environment John M. Plunkett,
Kathryn L. Plunkett		 This paper discusses the authors' experiences using SAS on VAX mini-computers as a teaching language for first-time computer students in a university environment. The sas base, macro loop source 4
205 Use of SAS Procedures for Estimating Design Based Logistic Regression Variances by Balanced Repeated Replication (BRR) Daniel H. Freeman Jr.,
Philip J. Leaf,
Martha Livingston Bruce		 logistic regression, sample, regression analysis, logistic distribution, proc calis, confidence interval, random sampling, standard deviation, proc step, macro definition source 5
206 Use of Version 5 SAS Software Explicitly Subscripted Arrays to Perform Bioavailability Calculations Patricia L. Gerend Most high level programming langu ages utilize the array data structure, which allows the programmer to group several related variables into one variable name. With version 5, SAS Institute has introduced into all its operating environments an array structure more consistent with that of other languages by pro viding for explicit subscripting of arrays. This makes it possible to access an array ele ment directly by specifying in a subscript af ter the array name which element is being sought. Since arithmetic calculations can be performed on the subscript variable. several array elements in any order can be accessed in a single statement. This direct access ap proach makes SAS code considerably easier to write, follow, and modify, features which are very important in programs intended for re peated usage. This paper will demonstrate the advantages of using explicitly-subscripted arrays in the calculation of bioavailability parameters auc, do over source 2
207 User Interface Tools and the Design of Interactive Systems Bob Pulgino This tutorial is intended to present a conceptual model that has proven useful in planning and designing interactive production applications using the SASiI!l System. It recommends a framework for the design, development, and maintenance of such applications at a strategic level, one that addresses the usual concerns of the systems analystand manager: * minimizing the time required to design and implement the application; minimizing the time required to train the user to operate the application, and making users'skills transportable from one application to another; insuring that the resulting applications are easy to maintain by promoting the use of modem, structured programming principles. The first section of the tutorial is an overview of the characteristics or capabilities of interactive applications that are easily attainable using SAS software in most environments proc step, dataset, macro variable, sas library, macro put, macro facility, input statement, macro definition, catalog entry, macro conditional, error message source 10
208 User-Written Procedures in C Leigh A. Ihnen source 4
209 User-written Procedures in Version 5 Leigh A. Ihnen source 4
210 Using Formats to Enhance the Appearance of Your Reports Stephen R. Schulz proc step, proc format, proc print, proc freq, proc tabulate, data cleaning, format statement, input statement source 7
211 Using New Features of Release 11.5 of SYSTEM 2000 DBMS Sue Trueblood Rakes This paper briefly discusses Release 11.5 of SYSTEM 2000r DBMS for I BM mainframes operating under as and eMS environments. The topics covered include user- requested enhancements that were implemented in part due to recent SASware BalloW results, Genius enhancements that affect the interactive report generator, and operating system specific enhancements source 3
212 Using PROC NLIN to Develop Capture Ratios Lynn Chapman This paper summarizes a method to develop capture ratios. The following topics are included: the or191n and reason for capture ratios; gathering and clustering data; calculating capture ratios with PROC NLIN; and using capture ratios proc step, regression analysis, proc nlin, response variable source 8
213 Using SAS Software as the Focal Point in Developing Integrated Systems Alan A. Parrow dataset, management reporting source 7
214 Using SAS Software Graphical Procedures for the Observer Agreement Chart Shrikant I. Bangdiwala,
Hope E. Bryan		 proc freq, data visualization, cohen's kappa, out= source 6
215 Using SAS Software to Develop a Prototype for Determining Participant Profiles and for Assessing Potential Relationships Among Health Insurance Variables Ralph F. Catalanello,
Bruce W. Eagle		 A major function assumed by many of today's Corporate Human Resource Organizations is the provision of data processing application ser vices and developments for Benefits Depart ments. A topic of much interest to benefit professionals is the cost factor associated with maintaining the physical and mental well being of the workforce and their covered depen dents. Executives nationwide have a growing concern over the perpetually increasing pre miums and expenses of employee health insur ance. And with this concern comes a very real need for expanded and flexible health care data analyses and reporting proc step, regression analysis, predictor variable, proc genmod, correlation, proc freq, proc glm, anova, multiple regression, response variable, sample, data science, health insurance, stepwise selection source 6
216 Using SAS Software to Pass Data to IBM's Graphical Data Display Manager Rameshwar Ramotar Developing presentation quality graphics using SAS/GRAPH often requires more advanced programming skills. Using the Interactive Chart Utility (ICU) of the Graphical Data Display Manager (GDDM). non-programmers can produce high-quality graphics, but considerable time must be spent on data entry. This paper outlines a technique to pass data to GDDM using TSO CLlST, FORTRAN, SAS, SAS/FSP and SAS MACRO time series analysis, macro definition, missing value, macro call, proc step source 5
217 Using SAS Software to Provide Information Feedback to Subjects in a Long-Term Study Carolyn J. Oakley proc step, sas base, proc printto source 4
218 Using SAS Software to Quantify Dissimilarity among Environmental Samples John M. Rinehart This paper provides computational support for a data analysis technique proposed by Ramsey and Marsh in B1gmetrics 40(3) (1984). Their paper, "Diet Dissimilarity,. outlined a method of quantifying the relative importance of prey taxa in determining overall dissimilarity of diets between groups of prada tors, using a Bayes classification procedure. The method can be extended for use as a general dissimilarity index between any two groups of multivariate environmental samples. SAS code is provided to implement the dissimilarity calculations under a variety of sampling models. It is hoped that this will stimulate further research into the theory and applicability of this technique. TntrQdUQti OD Environmental research often generates data which are essentially multivariate in character, with each "sample" consisting of a vector of recorded values, one for each of a set of descriptive variables sample, histogram, proc step source 6
219 Using SAS Software to Write a SAS Program: A Practical Introduction and a Survey Processing Sample Barbara A. McCandless One of the most powerful features of SAS software is its data ~tep programming language which offers map.y of the features of standard high-level languages like PL/I. This language opened up the possibility of having a SAS program write another SAS program, which can be a very useful way of eliminating a lot of tedious programming in certain applications. By customizing the useT interface, so that once a set of input parameters in simple format has been specified. the SAS software can then take over the writing and execution of the program. This paper teaches by example how to overcome three hurdles in using a SAS program to write other SAS programs. Simple working examples are followed by a full-fledged example of a SAS system to translate survey data file specifications into a SAS program for creating a SAS file and performing range checks on the data. This, is part of a larger SAS system which relies heavily upon using a SAS program to write SAS programs for survey data processing dataset, proc format, proc step, audit trail source 4
220 Using SAS/GRAPH Software on IBM Mainframes with Coax-attached PCs Gregg Ledford source 5
221 Using SAS/GRAPH Software to Display Box-Whisker Plots Charles B. Scott BOx-whisker plots are graphical rep~e sentations of data, primarily used in exploratory data analysis. They dis play the median, the box which spans 50% of the data, and the whiskers which extend to approximately 95% of the data; any points beyond the whiskers are outliers. The algorithm used here is a modified version of Tukey's (1977) design. The ANNOTATE facility is enlisted to plot the boxes, while PRoe GPLOT is used to set up and label the graph. This procedure puts the display in the user's hands and allows for flexibili ty of graph size, axes, width of boxes, and information contained within the display outlier, box plot, proc univariate, proc step, confidence interval source 6
222 Using SAS/GRAPH Software with a Protocol Converter Mike Kalt source 6
223 Using SAS/IML ""Software - A Multivariate Control Chart Example Phil Cirulli The intent of this paper is to show how to use SAS/IML software through the use of an example. Multivariate control charts for both location and variability is used as the example. Fully-documented code is shown utilizing many of the facets of SAS/IML including graphics, reading SAS data sets into IML, functions and printing of proc step, covariance, sample, proc iml source 3
224 Using the SAS System with Another Programming Language Frank R. Fry,
Hank Leddon		 An occasion may arise where data reside in a data base that is not easily access able to base SAS software. One solution to this difficulty is to create a fIle using a language that can readily access the data base and then use base SAS software to process the data in the file. This paper will discuss how this two step process can be accomplished. We illustrate with a pro gram used at the Federal Reserve Board to generate an alphabetical listing of depository institutions. The program uses an in-house pro gramming language to retrieve data from an IBM IMS data base and then uses Version 5 of the SAS language to process the data. It Is run on an IBM OS system input statement, proc step, proc sort, sas base source 3
225 Using the SAS System with Relational and Hierarchical Data Robert M. Hamer This paper is derived from a tutorial, in which I pre sented tools for reading and processing hierarchical and relational data using the SAS System. As such, the fig ures used are the transparencies I used in that tutorial, and this text is a brief explanation of each of those fig ures. This paper presumes some elementary familarity with the SAS System, expecially the language used in the DATA STEP. Specifically, it presumes familiarity with the use of the INFILE and INPUT statements to read data external to the SAS System, and to create SAS datasets. It also presumes some elementary familiarity with basic principles of data processing. Anyone with a few months experience using the SAS System to process data should meet these requirements. The first portion of this paper discusses the logi cal structure of a SAS job, and the activities that occur when a DATA STEP is compiled, and when it is executed dataset, input statement, proc step, proc print, if-then-else, data step, pointer control, structured data, trailing @ source 8
226 Using the SAS/C ""Compiler on the IBM Mainframe Oliver Bradley source 3
227 Using the SAS/GRAPH Software Interface to GDDM Martin Mincey This tutorial is for those users who use the SAS/GRAPH@ soft~ ware interface to IBM's GDDM (Graphical Data Display Manager) product. SAS/GRAPH software provides a set of device drivers that use GDDM routines to display graphics on terminals and to prOduce hardcopy graphics output on printers and plotters. This tutorial covers the following sections batch job, error message source 6
228 Using WINDOW and DISPLAY in the SAS System for PCs for the Patient Registry of the VA Coronary Artery Surgery Study Gary R. Johnson A system of programs was designed for the SAS System under PC DOS to update and edit a data library for a registry of patients from the Veterans Administration randomized study of coronary artery surgery. Data from yearly contacts with patients reporting cardiac events, anginal symptoms, medications, and hospitalizations are entered and edited in the system using customized window modules. The goals of the project were to implement the data management for the registry on the microcomputer while maintaining the capability to update mainframe SAS files. and to provide a user-friendly means for updating and monitoring the registry_ The design of the system. experience with WINDOW and DISPLAY statements. and an evaluation of using the SAS System Under PC DOS for the application will be presented data governance, data step source 5
229 Utilization of SAS Software to Solve Transportation Problems Jack Hudson Jr.,
Ginner Weatherby		 proc step, proc transpose source 6
230 Variance Component Estimation with the SAS System Ramon C. Littell,
Barbara G. McCutchan		 anova, likelihood function, restricted maximum likelihood, proc step, proc genmod, maximum likelihood estimation, proc mixed, fixed effects, proc glm source 6
231 VISUALS: Experimental Software for Dynamic Hyperdimensional Graphics Douglas P. Kent,
Warren F. Kuhfeld,
Forrest W. Young		 PROC VISUALS is designed to help you visually discover and formulate hypotheses about structure in multivariate data. PROe VISUALS does this by constructing a window into your multivariate data. Through the window you see a three-dimensional (3D) cloud of objects which represent your multivariate observations. You can smoothly spin this cloud in order to better understand its 3D structure. You can also move the window throughout the high-dimensional (hD) space, giving you insight into your data's hD structure. You do not use numbers, equations, or programming statements to control PROe VISUALS. Rather, you use single key-stroke, highly interactive commands. In addition to introducing PROe VISUALS, we review previous visual exploratory data analysis techniques, and we suggest future directions proc step, proc princomp, data science, exploratory data analysis, data analyst, multivariate analysis, proc cancorr, correlation, proc corresp source 10
232 What Good is the SAS System for Personal Computers? J. Philip Miller,
Robert P. Parks		 The SAS System for Personal Computers allows an IBM PC/XTjAT or compatible to execute the SAS sys tem. Many users see this as a cost effective way to use SAS compared to using SAS on a shared (mini or main frame) computer. We discuss this issue pointing out what costs should be considered and present some ex~ amples comparing the costs of a PC compared to a shared computer usage rating, quality of service source 8
233 What's New from the Education Division at SAS Institute Inc. Lee H. Evans,
Ann Lehman		 source 4
234 Writing Customized Reports Using the DATA Step Howard Levine file statement, data step, dataset, put statement, macro variable, by group processing source 10
235 XMERGE: A Generalized Macro for Subsetting and/or Expanding SAS Data Sets Thomas E. Schneider,
Donald D.M. Tong		 An efficient macro program XMERGE was written to merge two or more data sets ina manner similar to MERGE except that the resulting data set contains the cross product of observations, i.e., all possible combinations of observations across data sets. In the case where a BY statement is used, the cross product is generated within each BY group designated according to the BY variables. This allows a data set to be subset and/or expanded by merging observations according to the structure of another data set by group processing, cross-sectional data source 5
236 XRECALL: An Enhanced Replacement for the DMS RECALL Command Robert A. Frank XRECALL is a SASs Display Manager system (OMS) command macro. It serves as a functional replacement for the DMS RECALL command and adds the capability of recalling more than one or all submitted statement blocks at once macro definition, macro call source 6

Back to Conf go 1988 go 1986

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.