HOME QUICK SEARCH AUTHORS KEYWORDS HELP BUGS

NESUG 1995 Papers - Index

148 papers. Source: NESUG (Northeast SAS Users Group) conference proceedings.

Back to Conf
go 1996
go 1994
# Title
×		 Authors
×		 Abstract
×		 Keywords
×		 Extras Pages
1 A Creative Way to Read Stored Printed Reports Using the INPUT Statement and PROC TRANSPOSE Karen E. Davis input statement, dataset, proc transpose, raw data, unique key, proc step 2
2 A Data Warehouse Application under UNIX Cynthia A. Stetz Any data warehousing application requires managing the flow of data into and out of the data warehouse. This paper will discuss how we manage the flow of data from multiple sources to a SUN workstation running UNIX, and what we do with it once we get it there. Also discussed will be the methods used to increase the value of the data by creating summarized and denormalized views, the distribution of specialized data subsets for decision support, and an archival system designed to maintain the vitality of the warehouse by offloading aged data. Finally, future plans for enhanced exploitation and expansion of the data warehouse will be presented. History The Quality Assurance Reporting & Systems Group supports the data requirements of business analysts both within the QA Group and outside the group in other Merrill Lynch departments. The thrust in data support is towards allowing direct data access by the analysts, without the requirement of programmer intervention for every data request quality assurance, data warehouse, proc step, dataset 4
3 A Flexible and Handy Method of Custom Report Writing Sam X. Ye The DATA _NULL_ reporting step and the PUT statement provide the most basic and powerful code for SAS report writing for the pharmaceutical industry. Because of the diverse and complex nature of custom reports, the basic code to generate reports is usually lengthy and troublesome. This paper introduces a general macro called MFORM for generating custom data listings and statistical report tables, assuming the data set has been created. In the macro MFORM, coding to produce a table is similar to completing a simple fonn. Basically, only the table specific information is needed: tides, column headings, footnotes, variable names, and formats. The macro MFORM has a few easily understood parameters, comparable syntax with the PUT statement, and comparable functionality to the DATA _NULL_ step. The concept, structure, and components of MFORM will be discussed. Usage examples and their output will be shown put statement, macro call, data step, macro variable 7
4 A Further Introduction to SAS-L and Other INTERNET Resources David G. Carroll Anyone can share in the collective SAS" knowledge of users from around the world. Subscribing, posting, and searching on SAS-L are described. SAS Institute offers a variety of forms of electronic communication and adds new ones frequently. How to use these is explained. A sampling of ways to explore the Internet is also covered. The Internet The collection of networks that make up the Internet has perhaps 30 million users. That number, the tools they use, and the services and information they seek are in constant transition. Any attempt to describe this world, like the present paper, is doomed to being out-dated before the ink is dry. Nevertheless we will look out the window of this speeding train and gather some information about where we are and where we're going. What is SAS-L? SAS-L is, first of all, a BITNET list 6
5 A Generalized Method of Finding and Using Program Names John J. Genzano dataset, macro variable 2
6 A Generic Data Dictionary for SAS Applications Ralph W. Leighton data dictionary, dataset, proc step, proc contents, variable, missing value, proc print 11
7 A Guide To Including Logos and Pictures In Print Files Using SAS Mary A. McDonald,
Ernest Brandt		 macro definition, file statement 3
8 A Health Care Atlas of New York State Using SAS/GRAPH Michael S. Zdeb proc step, macro variable, bar chart, dataset, proc greplay, data step, proc sort 8
9 A Look At SAS Files Rick Aster dataset, libname statement, data step 2
10 A Macro That Generates the PROC GLM LSMEANS Table from the LSMEANS Output Data Set Michael A. Walega proc genmod, response variable, proc glm, proc step, proc calis, covariance, macro variable 2
11 A Macro to Manage Use of the Stored Program Facility H. Ian Whitlock,
Gerard T. Pauline		 The Stored Program Facility (SPF), introduced in version 6 of SAS, allows the user to save a DATA STEP in a compiled, executable form. For applications requiring the repeated execution of the same DATA STEP, the executable module results in efficiencies that can be seen in CPU and Elapsed (execution) times, and I/O. In each phase of the development, 1) testing the SAS code, 2) compDing the SAS code, and 3) executing the compiled module, the SPF user encounters the need to change his/her SAS code. This article presents a macro, SPFMGR, to manage these changes and execute stable code as the phase requires. The user need do no more than change one macro variable in passing from one phase to another. Hence, the code can be easily controlled by the submission process (e.g.: OS/JCl, CUST, DCl, SAS/AF, etc.). Our macro is based on an earlier more complex solution for MVS, presented in (1) by Neal Zimmerman, Observations (first quarter 1995) macro conditional, macro definition, macro put, macro variable, macro quoting function 9
12 A Macro Which Transposes Data Beyond what PROC TRANSPOSE Can Do Yen L. Ouyang proc step, character variable, proc transpose 2
13 A Macro-Based System for File Management and Version Tracking in SAS for Windows Release 6.10 Jason Blood Small programming projects that employ more than one person require simple and effective organization tools. Even modest programs that go through many versions and pass through many hands can cause mass confusion if they are not well documented. Statistics Collaborative. Inc. relies on a macro-based system for managing SAS programs, output, and logs. The features of the system include simplified file saving, version tracking, and output documentation macro variable, sas log 1
14 A Many-to-Many Merge of Recipe Files Using PROC SQL Adeline J. Wilcox,
Jean A. Fowler		 proc sql, proc step, order by 2
15 A Methodology For Supporting Summary Selectivity Dennis J. Clague outlier, sample, normal distribution 7
16 A Picture is Worth a Thousand Numbers -- Reporting Results of Longitudinal Analyses Karol H. Katz Univariate and Multivariate analyses of groups over a specific time period can be presented in tables or figures. Using SAS/GRAPH to illustrate comparisons often has an immediate impact and helps readers absorb the results of a study quickly. This paper offers both graphic and tabular displays of statistical comparisons between and within groups. The output is generated using PROC UNIVARIATE, PROC REG, PROC MIXED and displayed using PROC TABULATE and PROC GPLOT proc step, regression analysis, proc gplot, proc tabulate 8
17 A SAS Documenting Tool LeAnn P. Weaver Seeing the need to make documenting runs of SAS programs more streamlined, I developed a SAS program to assist me in my documentation. Instead of saving output run after output run, I now save one or more sheets of a report generated from the documenting program. There is room on the report to make notes of explanation and the purpose of the program. The reports can be bound and placed in a binder to document the flow of the programs. They can be easily followed for rerunning production runs in different cycles. The result is an easy way to document my work cpu time, dataset 2
18 A SAS Macro for Condensed Frequencies Bruce Ellis SAS~ provides extensive support for the examination of both numeric and character data. However, some of it's procedures offer pre-determined formats that can be clumsy and unnecesarily large. This presents two programs that use the SAS macro facility and the PAGESIZE printing option to generate multi column condensed frequencies for character or numeric data. The number of columns can range from one to more than a dozen and is determined from the length of the variable and the page width. The programs can run under MVS or MS-DOS, and require only base SAS macro variable, proc freq 3
19 A SAS Macro Introduction to Statistical Quality Control of Longitudinal Data Timothy D. Kelly This paper will describe a macro created by IMS America's Statistical Services Department to help control the quality of data for the Electronic Medical Records Project. I will briefly explain the structure of the data, the implemented methodology, and the macros that were developed to check longitudinal data quality at a physician practice level moving average, proc step 2
20 A SAS/AF FRAME Application for Clinical Laboratory Review Denis Michel A major component of the safety review of a clinical trial is the analysis of laboratory data. Standard tabulations and listings have been used to provide clinical research staff with the infonnation required to determine the safety of a therapy. However, custom reports are often required in order to meet the needs of various therapeutic areas and different reviewers. For example, a reviewer may want to focus on a particular subject population, or a certain group of laboratory tests, or may want to define laboratory abnonnalities in unique ways. To accommodate the needs of the reviewers, a SAS /AF FRAME application was developed to enable clinical research staff to review laboratory data. The users have the capability of analyzing the data for a clinical trial without assistance from programmers and statisticians. The system features are described as well as the data flow, SAS data sets, and programming techniques used in building the application laboratory data, clinical trial, scatter plot, where statement, outlier 7
21 A Simple Methodology for Developing Clinical Reporting Macros Paul M. Gilbert,
Steven Light		 data cleaning, adverse events, proc step, proc report, macro parameter, csr, clinical trial, statistical programmer 6
22 A Survey of Capability Invoices Dennis W. King confidence interval, proc step, histogram, sample, exponential distribution, normal distribution, proc print, maximum likelihood estimation, likelihood function 11
23 A Truly Integrated Method of System Design Dennis J. Clague The purpose of this paper is to describe a macro variable, keep statement, macro definition 7
24 Advanced Features of PROC REPORT in Batch Mode Christopher Yindra Proe Report combines the power ofProe Tabulate with the flexibility of DATA STEP reporting. This paper will clarify how Proe Report constructs a report and will present some advanced reporting techniques available with Proe Report. Review Basic report syntax: ··BegIn report in batch mode···* PROC REPORT DATA=CYLm.EMPLOY NOWD HEADLINE HEADSKIP SPLlT='·'; proc report, proc step, data step, character variable, numeric variable, proc format, missing value 10
25 An Efficient Way to Condense Unbalanced Panel Data Andrew J. Epstein unique key, numeric variable 2
26 An Introduction to PROC FSEDIT and PROC FSVIEW Alan C. Wilson rhe FSEDIT and FSVIEW procedures in SAS/FSP software allow for interactive, real time maintenance and modification of SAS data sets on a record level. This tutorial will discuss data set creation, browsing using interactive WHERE conditions, and rnocflfication including adcfmg, updating and deleting observations. Data entry screen modifications in PROe FSEDIT will not be covered, however the use of formulas in PROe FSVIEW to interactively create new variables and modify existing variable values will be illustrated proc step, where statement, character variable, dataset, libname statement 7
27 An OS/2 Help File for SAS Tracy L. Lord This paper describes a documentation sas programmer, dataset 4
28 Answers To Commonly Asked Statistics Questions Mike Stockstill The purpose of this paper is 10 address some common "how 10" and ''what does this mean" users' questions collected by the statistics group in Technical Support. Product discussion includes SAS/IML·, SAS/INSIGHT", SAS/STAT", SAS/OR', SAS/QC", SASJETS", SASlLABo, .IMP", and base SAS' software. The emphasis is on what can be done with the current rele.ases rather than with future releases, although comments are provided where appropriate. Understanding the paper requires little or no technical or statistical expertise. The paper should be mOSI useful to site representatives who face a lot of stat· oriented questions. Topics addressed are alphabetized by procedure as much as possible, and a miscellaneous section follows at the end. The paper itself is based 011 our internal notes to each other and is Dot really intended to stand alone as any sort of documentation. We all contribute, and information is constantly updated 19
29 Appendix Numbers: A Macro to Eliminate Hard Coding Daphne E. Ewing When a project requires appendix numbers to be displayed on each of many output displays, and there is a possibility of them being modified, even once, hard coding them into each program makes a change of this nature time consuming and tedious. Within the report code, be it DATA JWLL_ or PROC REPORT, a macro call can dynamically generate the appendix number making this dynamic portion of the display easy to maintain. Identifying Report Assuming each report generated has a unique appendix number which potentially could change, the program which generates the report has to identify this report. Linking the program name to the appendix number makes for an easy cross reference. Using a %LET statement in each program that coincides with the program name allows for decoding through a format statement to generate an appendix number. The PROC FORMAT code should reside in a separate me. Below is a sample for defining appendix numbers which should be placed in the format library proc report, proc step, proc format, macro definition, %if-%then-%else 2
30 Batch vs. Interactive - Why You Need Both Janet E. Stuelpner On the mainframe, a great number of people write their programs and then use batch submission exclusively to run and test them. As we all know, at times this can be very frustrating. Others use interactive sessions exclusively. This paper will show the user the reasons why a combination of using batch submissions along with an interactive sessions are beneficial to developing new programs batch job, batch processing 5
31 Beginning Tutorial SAS/ASSIST; Explore the 'Point-and-Click' Approach Evelyn F. Myers SAS/ASSIST can be a valuable tool for the new or seasoned programmer. 1bis software allows the new user to access the power of SAS without teaming to program. Occasional users do not have to remember the rules of SAS syntax. The experienced user can test applications quickly, then save the code that is generated from SAS/ASSIST to modify later. 1bis menu-driven software canhelp SAS users perform such evetyday activities as generating reports, summarizing data, developing graphics, accessing data in external files, and editing and submitting programs. The only requirements for learning SAS/ASSIST are to know how to log onto the computer system and how to begin the SAS session. The Primary Menu and several related menus will be explored concentrating on Data Management, Report Writing, Graphics, and Data Analysis. A list of the SAS system packages needed for SAS/ASSIST to be fully functioning will also be provided where statement, proc freq 5
32 Building a Data Warehouse with SAS Software in the UNIX Environment Karen Grippo,
John Chen		 A common trend in corporations today is "downsizing" - moving applications from expensive mainframes to more affordable micJ.range platforms. The Information Center of Dun & Bradstreet Information Services, faced with the challenge of moving fulfillment off the mainframe, has chosen SAS software not only as the vehicle for its data analysis and delivery but also as its database management system. Twenty gigabytes, extracted from mainframe legacy databases, have been loaded onto an HP-UX platform as indexed SAS data files. Two data structures (a table mapping the data files to file systems and a data dictionary mapping data elements to data files) and a set of SAS macros have been developed to give the Information Center developers and business users transparent access to the data, freeing them from the need to know data file or data element locations dataset, macro definition, proc sql 5
33 Capitalizing Names With PROC FSVIEW Formulas Alan C. Wilson If the job you are working on contains character data in the wrong case, it is easy (in SAS versions 6.09 on) to use PROC FSVIEW formula commands to create the appropriate modifications interactively binary variable, proc step 2
34 Converting PROC FREQ Output to a Data File Charles A. DePascale,
Patricia Cole		 proc freq, proc step, dataset 2
35 Cooperative File Interfaces: How SAS & BMDP Can Work Together On VAX/VMS Thalene T. Mallus Addressing compatibility problems between software products is common among data managers and analysts. This paper will discuss the flexibility of the SAS System to interface seamlessly with other file formats and statistical packages under the VMS operating system. Specifically, SAS's ability to interface with BMDP, a library of statistical analysis programs developed at the UCLA Health Sciences Computing Facility, will be examined. Code examples for the use of the BMDP interface library engine, PROC CONVERT, and an overview of the omitted VMS SAS documentation for PROC BMDP will be discussed. Advantages and disadvantages of each interface method will be explored proc step, dataset, proc copy, data governance, character variable, libname statement, select, statistical software, sample, by group processing 10
36 Create a Permanent View by Querying Data From Two Tables Owen Jiang,
Robert Musterer		 To maintain a subset from two members requires additional space in a permanent library, and also an update process if any of the members has been changed. Creation of a permanent view against the two members could make the maintenance automatic and save the additional space. An example case wiII be described from a clinical trial data base where both study drug schedule and antimicrobial data members are created in the permanent library. To analyze the antimicrobial drugs used at the different periods, three subsets (previous, during and post treatment) have to be created. Creation of permanent views instead of permanent subset members could save space and maintain the views automatically against both the study drug schedule and antimicrobial data members. Case Study: In a common clinical trial SAS data base at BAYER Corp., we usually have a study drug schedule member which contains study drug start and stop dates for every patient proc step, proc sql, proc sort, clinical trial, create view, order by 3
37 Creating a Bar Chart with a Moving Line Target Without Using Annotate Rick Guest GPLOT is normally used to create line graphics. This paper describes how SAS GPLOT is used at the IBM Microelectronics Division Manufacturing facility in Essex Junction, Vermont, to explode the size of the needle plot, thereby making it appear as a bar on a graph. This, in turn, makes it easy to add another variable name and plot that name as a line. An example is presented that enables programmers to create a bar chart quickly and with little effort. USING pRoe GPLOT When programmers typically think about a bar chart they focus on PROC GCHART. This paper describes how on some occasions PROC GPLOT can be used to create a bar chart. First, the data required (below) is placed in a SAS.WORK file called SAMPLE bar chart, proc step 2
38 Creating ASCII Fixed Length Record Files from SAS Data Sets Lawrence W. Altmayer dataset, proc freq, data step, proc step 2
39 Data Reduction: An Introduction to Principal Components Analysis Jennifer L. Caveny,
James F. Murray		 This paper is an introduction to the method of Principal Components (PC) Analysis and the SAS PROCEDURE PRINCOMP. First, we will give a quick ovelView of the proc princomp, correlation, covariance, proc step, missing value, standard deviation, variable list 10
40 DATA Step Reporting Marge Scerbo put statement, file statement, set statement, by group processing, proc step, options statement 7
41 Developing an SCL End-User Automation System Michael B. Moll,
Alan W. Bradt,
Angie Lymberg		 From 1m to 1994 the Municipal Health Services Program (MHSP) rLthe Health Care Financing Administration (HCFA) has been processing 431,000 forms annually. This is a processing rate of approximately 1,740 forms per day from four cities. In the past the data has been received via a computer tapes or on paper forms. The data was processed using a ten year old COBOL program that was user initiated weekly or biweekly in batches. The data submitted on the forms was validated, verified and edited. Accepted information was included in monthly HCFA reports and unacceptable data was returned to the sender for correction. In addition, the current form (HCFA 127) needed to be replaced by the National Health lnsurance Claim form (FICF A 15(0). This was done so that the Demonsiration staff could gather more detailed health information for administration purposes. In July 1995 the batch system was replaced with a menu driven end-user SAS SCL system. The SAS SCL menu system proc step, proc append 6
42 DIFFER: An SCL File Comparison Utility James R. Carter catalog entry, proc step 4
43 Displaying the Results of Multivariate Analyses Dan Jacobs,
Steven G. Smith		 Several multivariate statistical techniques, such as principle component analysis, orthogonally transform variables into linear combinations. The resultant axes are commonly used to displayed the transformed data in a lower dimensional space. Interpretation is improved when related to the original variables (e.g. biological, environmental, spatial, temporal). This paper demonstrates how to use SAS· to partition multivariate structure of the original data into a series of plots that can be used to interpret multivariate analyses such as PCA proc princomp, sample, scatter plot, proc step, correlation, if-then-else 8
44 Distributed Processing: Using Remote Library Services to Bring Storage to the Workstation Daryl R. Hoffman proc sql, distributed processing 5
45 Dynamic Data Selection Lists in SAS/AF Programming Entries John M. Rinehart SAS applications often involve reporting or graphing a subset of data from a large database. Generally, the user is asked to provide data selection or screening criteria used to identify the desired subset. In addition to fixed choices ("this year I last year", "Division A / Division B", "sales I profits", etc.), these criteria may be open-ended and based on actual data values contained in one or more columns of the database ("report for these SKU's ... H). The easiest and most reliable way to obtain such values is to have the user select them from a list of available values. This paper presents one method for implementing data selection lists in a SAS/AF program screen with SCL, by querying the source dataset for available values. The proc sql, create table, sample, missing value 7
46 Dynamically Build a PROC FORMAT for Subsetting Large Datasets Warren E. Stinson proc step, proc format 1
47 Effective Use of SAS Select Language Statement Rick Lockwood IF-THEN statements are commonly used to test conditions. Many times, long pages of IF-THEN statements can slow down code processing time. However, there is an alternative, the SAS SELECT Language Statement. Using SELECT statements when a long list of mutually exclusive conditions exist can make code more efficient, legible and fun to write. This paper will discuss how easily SAS SELECT can be put to use. Description The SELECT statement is commonly used in place of IFITHENIELSE coding in SAS. It enables one of several statements or multiple statements to be executed. legibility When we, as coders, write short SAS routines to accomplish some specific task, it probably doesn't matter which technique is used for decision-making statements. However, as routines become longer and more complex, technique becomes more important. Most programmers who have tried to decipher their own code after a period of time know that clear, concise and well organized statements make coding much easier select, if-then-else 2
48 Efficient Data Transformations Using User-Defined Formats and Informats Kim L. Kolbe Ritzow Regardless of your level of experience, this poster can help you understand the efficiencies that can be gained by transforming your data through user defined formats and informats using PROC FORMAT. It will make a comparison with other data transformation methods like IF-THEN and SELECT statements in an attempt to show how user-defined formats and informats are a better means for data transformations within our SAS programs proc step, select, proc format, dataset, data transformation, flat file, if-then-else, user-defined format 4
49 Empowering Researchers: Using SAS/AF to Create A Graphical User Interface to SAS, Sybase, and the UNIX System Charles Perdue Sutelan proc sql, proc freq, view, dataset, create view 9
50 Estimating MEXVAL Statistics Ashok Mehta regression analysis, proc calis, proc step, predictor variable, response variable, estimate 5
51 Estimation of Nonlinear Models: A Computational Review D. Das Purkayastha proc autoreg, time series analysis, regression analysis, maximum likelihood estimation, proc calis, likelihood function, response variable, proc nlin, proc forecast 8
52 Ferocious Functions Tamed John D. Shuey sysdate9, sas function 5
53 Frame Your View of Data with the SQL Procedure Kirk Paul Lafler The SQL view is a powerful feature of the relational model. Data becomes visible as a framed picture through which you can see information. This paper presents the many features of an SQL view and how it can be used as a building block by users and application programmers alike. A tutorial approach will be used to illustrate the virtues found in building, using, and exploring the numerous practical applications of viewing data proc sql, order by, select, create view, group by, proc step 4
54 Gaining Efficiency With SAS Software Kirk Paul Lafler With the ever increasing popularity ofSASI' Software among programmers and users, guidelines for its efficient use become increasingly important. This paper presents a collection of techniques for gaining efficiency when using SAS Software. Areas deserving special consideration include program execution, I/O, disk space as well as an often overlooked item known as program maintenance. Sample code is presented to illustrate numerous practical techniques as to why some proc sql, cpu time, dataset, drop statement 6
55 Generate and Organize Your Code into Windows Help Files with SAS Mark Penniston Windows help files allow a structured documentation vehicle to organizing programming code. By passing SAS code and creating topic links programmers can easily find. reference. and retrieve site specific macros and programs. With SAS's strong string and file manipulation features one can generate these help files in both batch and interactive modes. Features within these help files allow programs which link to macros and other files to jump to said programs with the click of a mouse button. In large corporate SAS environments this can be particularly useful where many programmers are developing multiple systems at the same time. By being able to look up macros and code easily the programmers can reuse code developed by others without tedious searching. This paper demonstrates basic features of such a system and practical usages of the resulting help files dataset, txt, macro variable, data step, macro definition 7
56 Getting Into a New FRAME of Mind Object Oriented Programming with SAS FRAME ENTRY Software Paul Butler The transition from traditional modular or structured program design to an object oriented approach is often a gradual one. Programmers not experienced in implementing applications using the OOP paradigm frequently continue to use structured techniques and slowly incorporate OOP strategies as they become familiar and more comfortable with them. It is likely many SAS Frame Entry programmers follow this trend. This paper will examine three different techniques used to perform the same task (employing the SAS Frame Entry), and then make comparisons as to the ease of their implementation and the degree to which they follow the object oriented model. Each method reviewed will use an increasingly object oriented approach, and in so doing illustrate the often deliberate transition when shifting from a structured programming methodology to an object oriented one sample, cpu time 10
57 Getting Organized in MAC OS Rick Aster options statement, filename statement 6
58 Getting Started with PROC LOGISTIC Andrew H. Karp logistic distribution, logistic regression, predictor variable, regression analysis, proc step, proc logistic, response variable, odds ratio, proc reg, confidence interval, null hypothesis, data science, estimate, proc genmod 5
59 Getting Subtotals and Percentages: Its Never Been Easy Christopher Toppe Getting subtotals and percentages onto a database has always been one of the curses of working with SAS data. The most common techniques are neither easy to conceptualize nor easy to program. Sorting your data and using FIRST. and LAST. variables is perhaps the standard brute-force programming methodology that we alI have learned to use and come to hate. Using PROC MEANS (or PROC proc sql, group by, proc step, by group processing, proc means, create table, sas base, dataset 6
60 Getting the Most from the SAS LOG Robert G. Hall merge statement, put statement, error message, input statement, missing value 4
61 HOTROD: An Intelligent Hotdeck Imputation System Eugene M. Burns,
Jay Olsen		 imputation, response variable, multiple imputation, sample, missing value, macro variable, macro definition 8
62 How to Put Almost Anything into a PROC TABULATE Cell Michael E. Tomb Complex table cell fonnatting within PROC TABULATE can be created by using the SAS Macro Language to produce on-the-fly fonnats. In the example to be discussed, the upper and lower 95% confidence bounds of the relative risk odds ratio are calculated from PROC FREQ for a series of adverse event comparisons. Both bounds are then displayed within the same table cell. The technique illustrated can be used whenever the value of more than one variable is to be shown within a single table cell proc step, proc tabulate, macro variable, confidence interval, proc freq, odds ratio, macro facility 2
63 Implementing a Modified Method of Weighted/Unweighted Means Robert H. Brown sample, sample size, numeric functions, treatment effect, weighted mean, output delivery system, regression analysis 9
64 Importing and Exporting SAS Output With Microsoft Word For Windows Edmond Williams,
Shi-Tao Yeh		 The SA~ system produces statistical output, tabular material and graphics output from different SAS procedures. The Issues of how to export SAS/GRAP~ output to word processing application software and the enhancements of output quality will be discussed sas output, proc step 7
65 Imputing Missing Values to Survey Data: Three SAS Algorithms Roald Euller imputation, regression analysis, missing value 2
66 Interval Training: Fun with Date Functions Alan Dickson sas base, macro definition 2
67 Introduction to Arrays Bob Virgile 5
68 Introduction to SAS/GRAPH John J. Cohen Opening the SAS/GRAPH manual for the first time can be an intimidating experience. Rudimentary charts and plots seem to require more study than a programmer can spare. This tutorial will lessen the effort required to take advantage of SASfGRAPH. The presentation begins by examining the relationship between the response data set and the more common SAS procedures used to create displays, such as GCHART, GPLOT, and GMAP. Customizing graphics using TIlLES, FOOrnOTES, and LABELS is also covered. The presentation concludes with a discussion of hardware considerations and an exploration of the ANNOTATE and GREPLAY procedures proc step, proc gchart, proc gmap, proc gplot, bar chart, sample 8
69 Large Data Volume: Tricks of the Trade Allison Levine While working with 30 million observations and a multitude ofvariables, one can encounter many problems. Your PROC SUIDJII3Iy abends, the code becomes so difficult to follow you need a road map, and variables seem to multiply on their own. Then, when you finally get everything wOrking, you get exiled to tape because your files fill half the DASD in the shop. What's a programmer to do7 This paper illustrates several techniques used by the author to handle problems associated with large data volumes. Several options for summarizing extremely large SAS files are presented, including how to optimize PROC SUMMARY's data capacity and when to use DATA STEP summarization. In addition. advanced methods for minimizing tape processing and simplli)ting code bloat using DATA STEP Views, Arrays, and Macros will be discussed. Finally, we'll discover how to get more than one piece of information from a user-defined format. Data Views Resources are always an issue when dealing with large data sets proc step, by group processing, proc format, data step, proc sort, label statement, proc means 5
70 Let The SET Statement Do Your Record Counting For You Michael A. Walega conditional do loop, set statement 2
71 List-Less Code: Energizing Your FRAME Programs Using SCL Lists David Bromberg Screen Control language (SCl) Lists are a critical element of SAS/AF FRAME entries, objects, and widgets and can be used to enhance the flexibility and capabilities of SCl programs. Lists are data depositories that can contain an unrestricted mix of character and numeric data that can be extended to memory limit. Lists are critical to FRAMEs because they drive the display of objects. Lists are declared, accessed, transformed, managed, and displayed via functions. Lists process rapidly because they are manipulated in RAM. Understanding lists is important for capitalizing on the full capabilities of the SAS FRAME entry and the SAS/AF application environment proc print, catalog entry, proc step, macro loop, %if-%then-%else, macro definition, proc sql 9
72 Logical Analysis of Data (LAD): An Implementation In SAS Language Endre Boros,
Peter Hammer,
Victor Kamensky,
Alex Krogan,
Ilya Muchnik		 LAD is a new group of lllethods of elata analysis based on extensions of Partially Defined Boolean Functions * PDBF (CRAMA,HAMMER and lBARAKt,1988). PDBF is defined by a given set of observations, like training sample in machine learning, artificial intelligence and pattern recognition. PDBF extensions constructed by LAD lire used as decision-making rules for new observations. It seems at a glance that logical (combinatorial) flag variable, fisher exact test 6
73 Macro Based Reshaping of Pharmacokinetic Data for Crossover Studies Doreen M. Turczyn-Toles During the e.lTly stages of drug development. study monitors need to assess the body's absorption. elimination and tolerance of study drug over time. To meet this need. programmers must often reshape pharmacokinetic data to produce both tables by time and graphs by subject. Programming quickly becomes complicated when time profiles need to be differentially fonnatted by treatment leg and by unequal number of subjects. The following program is designed to allow program flexibility while customizing the results for the individual study monitor by using SAS macros and macro variables proc step, macro variable, proc gplot, standard deviation, proc calis 9
74 Maintaining Dynamic Permanent SAS Format Catalogs Across Mainframe and Personal Computer Systems Helen-Jean Talbott How do you get everyone in your department to group the same data in a consistent manner and to translate codes into the same text on reports regardless of whether they are using the mainframe or personal computer for their work? The answer lies in developing user-defined formats, storing these formats in a permanent SAS format catalog, and maintaining this format catalog in parallel on the mainframe and personal computer systems. These tasks are relatively straight forward under SAS version 6 procedures FORMAT, CPORT, CIMPORT, along with SAS/CONNECT software. Benefits of this approach include ease of maintenance, conSistency across programs by multiple users, improved program efficiency, and enhanced report appearance. SAS versions discussed for this paper are SAS version 6.08 under MVS for the mainframe and SAS version 6.10 under Windows for the personal computer user-defined format, proc step, proc format, format statement, dataset, proc cimport, proc cport 5
75 Managing Large Databases With Customized SAS Windows Ronald W. Matheny This paper discusses the principles of database management through customized windows using SAS!AF, particularly PROC BUILD, to invoke interac tive and batch processing of data entry, editing, updating, automatic report generation, and custom report generation functions, including graphics. SAS! ACCESS is used to efficiently store the data in ORA CLE using about a third of the memory that SAS databases require. Customized "views· are devel oped using PROC Sal with PROC/ACCESS and BASE/SAS to efficiently manage extraction of the data from the ORACLE database into SAS for a vari ety of purposes. Security from both the ORACLE and SAS sides is discussed as it relates to a network of personal computers and Unix workstations and Unix servers including storage of data on different Unix machines to allow more effective controlled access to the data quality assurance, quality control, sample, missing value, proc step 10
76 More Advanced Features of PROC REPORT Kim L. Kolbe Ritzow This paper is designed for users who have an advanced knowledge of PROC REPORT's syntax. The paper wUl focus on uses of the FLOW option, the creation of multi-column reports, customized report breaks, and creating reports containing percentages proc report, proc step, character variable, numeric variable, put statement 6
77 More on PROC PHREG: Conditional Logistic Regression (or Should We Do Another Study..?) Peter P. Gaccione logistic distribution, likelihood function, treatment effect, proc step, maximum likelihood estimation, logistic regression, regression analysis, proc phreg, null hypothesis, proc logistic, conditional logistic regression, response variable 10
78 Multiple Minima: Phenomena of The Data Or A Drawback Method? (Comparison of SAS/IML NonLinear Optimization Routines) Ekaterina Gibiansky Several SASIIML nonlinear optimization routines and the least squares routines were used to estimate the parameters of a nonlinear model. The model came from a pharmacokinetic setting, and had a complicated, highly nonlinear form. Eight parameters had to be estimated from 73 data points for each of ten data sets. The derivatives of the objective function were not provided analytically. The analyses were repeated with 100 random (in a plausible range) sets of starting values for each of the ten sets of data. The only two methods that consistently ran without error messages were the. Neider-Mead Simplex Method (NLPNMS) and the Quasi-Newton Method (NLPQN) estimate, sysrc 4
79 Multivariate Density Estimation using SAS Software Jeffrey S. Butler histogram, sample, k-nearest neighbors, proc freq 9
80 Novel Uses of Extended Tables in SAS-AF Program Entries James K. Schwar Extended tables can be used in SAS-AF Program entries as a means ofbrowsing data and also as selection lists. This paper will discuss the programming techniques used to create a SAS-AF application in which extended tables are used not only as a browsing tool and as selection lists, but are used as a way to display data in the form of a Pareto chart, with the data swnmarized by any variable chosen by the user. The applicaticn serves as a tool whereby the user can "drill down" through the data with complete flexibility as to which variables are used to accomplish the drill down, and in which order. This application is very popular among end users at Carpenter Technology and could be adapted easily to work with anydataset which lends itself to a drill down approach for data analysis bar chart, proc freq, proc step, dataset, character variable, if-then-else, temporary dataset 7
81 ODBC: Windows to the Outside World S. David Riba,
Elisabeth A. Riba		 proc sql, excel 10
82 OLE and the SAS System for Windows Release 6.11 Jennifer Bradshaw Clegg,
Mark W. Cates		 This paper describes the support for OLE (object linking and embedding) provided in the SAS System for Windows Release 6.11. Features include support for embedded and linked objects, drag and drop, visual editing, OLE automation, and OLE controls. OLE support within the SAS System is available in SAS/AF FRAME entries and SASIEIS applications excel, catalog 8
83 Performance Management in an Open Systems Environment Using SAS/CPE with SunNET Manager Joseph M. Mirabal The effects of downsizing Legacy Systems has resulted in a significantly increased demand for UNIX based tools to conduct analysis and reporting of data processing performance. Legacy systems supplied us with a centralized repository of information that is best described as comprehensive. There is little of this in UNIX based systems. The approach presented in this paper is one that follows in the heritage defined by analyzing the data center performance translated to the specific requirements of client/server technology. Providing a Performance Management Methodology for any client/server applications environment presents a unique challenge to the analytical science of computer measurement. The distributed nature of Client/Server technology provides us with managing a multitude of processing centers where the data can be processed. In the Legacy Systems environment the nucleus of all processing was the mainframe computer which over the years grew to handle increasingly more work network utilization, sample, cpu time 16
84 Playing With the Mandelbrot Set John J. Cohen Fractals generally and the Mandelbrot Set in particular have offered a rich area for research and exploration for students of mathematics and graphics. A delightful result has been a plethora of dissertations and publications brimming with beautiful graphs. For SAS programmers, in contrast, these have provided us unwished for opportunities for testing the limits of our patience, our programming skills, and the hardware platforms on which we run. All too frequently the unhappy consequence of our dabbling has been headaches, programs which run for hours (or days!), and empty toner cartridges. There is no reason to let others have all the fun, nor to wonder about just what kind of arrangement made it possible for their plots to print. We will offer here the details of a successful experience in producing pictures, secure your first born, and list programs worth copying (assuming proper citation, of course) macro facility, sas base 3
85 Power Analysis: Prospective Use of a SAS Macro John F. Walsh The use of the POWER macro (Latour, 1992) for teaching statistics to graduate students in the behavioral sciences is illustrated. The stored macro capability is used to facilitate student use of SAS software. Students design experiments using the summary cell statistiCS of n, mean, and stan dard deviation. Effect size is conceptualized in terms of separation among cell means. The sum mary statistics are sufficient to perform a PROC GLM analysis (Larson & Wartenberg, 1989). Output from PROC GLM in the outstat data set is then used by the macro to compute power and sample size calculations for one-way and factorial designs. Conceptualizing effect size in terms of separation among cell means for main effects and interaction terms is a user friendly vehicle for translating classroom concepts into the features of actual experiments effect size, sample, proc genmod, proc step, main effect, proc glm, sample size 3
86 Practical Consdierations in Estimating Linear Dose Response From A Phase I/II Titration Study With Subjects As Random Effects Virginia Fitzpatrick,
Ruth Swithenbank		 This report describes and compares several proc mixed, proc step, likelihood function, maximum likelihood estimation, regression analysis, covariance, fixed effects, restricted maximum likelihood, estimate, repeated measures, proc reg 7
87 Practical PROC CALENDAR Andrew Mosso proc step, dataset, date variable 2
88 PROC FORMAT: Use of the CNTLIN Option For Efficient Programming Karuna Nerukar,
Andrea Robertson		 PROC FORMAT can be a useful tool for improving programming efficiency. This paper will demonstrate how to use PROC FORMAT with the Cntlin option to read in an extemal dataset and convert it into a custom format. The paper will discuss pitfalls to avoid when using the PROC FORMAT such as using a numeric variable containing a decimal as your LABEL. The paper will also demonstrate the use of PROC FORMAT as a more efficient mechanism than sorting and merging when performing a table lookup (i.e. using a key variable to extract observations from one dataset based on the values of the key variable in a second dataset) proc format, lookup, proc step, data step, proc sort, insurance claims, conditional do loop, cpu time 7
89 PROC REPORT - Capabilities and Techniques for Batch Users Larry Gertzog proc report, proc step, data step, proc means 9
90 PROC REPORT: An Introduction to the Batch Language Ray Pass proc report, proc step, sample, class variable 10
91 PROC SUMMARY: A Powerful Exploratory Data Analysis Tool Kim L. Kolbe Ritzow Starting with Version 6 of SASe Software, PROC proc means, proc step, dataset, class variable, flat file, output 10
92 PROC TABULATE: Controlling Table Appearance August V. Treff PROC TABULATE provides one, two, and three dimensional tables. Tables may be constructed from class variables, analysis variables, and statistics keywords. Descriptive statistics can be displayed in table format with a minimum of code. This paper will use a five.response survey to demonstrate how PROC TABULATE features translate to a printed report. We will examine table dimensions, crossing and concatenating variables, and percent calculations. Controlling table appearance through formats and customized column headings also will be discussed proc step, proc tabulate, proc format, proc means, class variable, label statement, proc freq 8
93 Producing Effective Graphs with SAS/GRAPH Software Betsy Corning 11
94 Rank Based Alternatives to Normal Least Square Analyses Maria C. Prenger sample, proc step, confidence interval, proc means, proc sql 6
95 Reading External Files Using SAS Software Clinton S. Rickards,
Michael L. Davis		 In the real world, most data are not stored in SAS data sets. Typical data repositories used by business and academia include legacy mainframe files, client/server relational database management systems, and simple text files on desktop computers. While we want to use SAS to examine, to analyze, to display, and to graph this data, SAS works best with data stored in SAS data sets, not with data stored in their native forms. Fortunately, SAS has very powerful tools for reading a wide variety of extemal files, from the most rudimentary to the most complex. This paper is an updated and expanded version of similar papers presented by the lead author at previous SAS conferences and will introduce DATA STEP techniques for reading standard extemal files not stored in data base management systems or proprietary file structures. . We will begin with a brief overview of what to do before writing any code, followed by an examination of the SAS statements used to read an external file dataset, length statement 10
96 Reducing LAN Traffic for SAS Data Sets: The Skinny File Approach Richard D. Langston,
Meade Morgan		 This paper describes an approach used by 1he Centers for Disease Con1rol to reduce the LAN traffic on a group of very large SAS data sets. This approach, Implemented as an engine using SAS/TOOLKIT Software, involves dMdlng a single $AS data set into many single-variable data sets. The user specifies which variables to access, and the engine handles 1he building of program data vectors consisting of only those variables. Also, each single-variable 'skinny file' can be further reduced in size using various compression techniques. Thus the engine allows for a very significant reduction in disk usage, network traffic and 1hroughput time dataset, cpu time, traffic data, proc step 3
97 Repetitive Data Retrieval from Time-Series Data Robert Charles Oshinsky macro definition, permanent dataset, time series analysis, macro variable, %if-%then-%else 2
98 Retrieving Data from Large DB2 Tables Based on Keys in SAS: The Sequel Judy Loren,
Jack N. Shoemaker		 data step, macro variable, where statement, macro parameter, proc sql, dataset, macro definition, call symput, character variable, %if-%then-%else, sas log, set statement, macro loop, create table, sysrc, proc step 10
99 SAS Code Coding SAS Warren E. Stinson 2
100 SAS Data Views: A Virtual View of Data John C. Boling The concept of a SAS data set has been extended or broadened in Version 6 of the SAS System. Two SAS file structures now function as SAS data sets: SAS data files and SAS data views. Both file structures can be processed as SAS data sets in DATA steps or pro cedure steps. This article introduces the concept of a SAS data view, presents Its advantages and disad vantages, discusses the three types of SAS data views, and compares its structure and processing with that of a SAS data file. Defining a SAS Data View The SAS data file is already familiar to existing SAS users because it represents what, historically, has been called a SAS data set. For example, 1982 Version SAS data sets and Version 5 SAS data sets are technically referred to now as SAS data files. This is essentially a change in tenninology. The new file structure and def inition in Version 6 is the SAS data view. Unlike SAS data files, SAS data views do not con tain actual data 7
101 SAS FSP in the Small SAS-only Shop Developing a Training Information System with Minimal Resources Floyd Schwartz dataset, proc step, data integrity 6
102 SAS Macros for Bootstrap Samples with Stratification and Multiple Observations per Subject Terry Hyslop The development of powerful computers for the desktop has allowed the bootstrap to become a popular inferential method in statistics. A bootstrap sample is a random sample selected with replacement. While SAS macros for selecting bootstrap samples have been pl'e!!ented, the issues of multiple, and varying numbers of records per subject have not been addressed. This data structure would arise, for example, in the analysis of mixed-effects models, such as for data collected from crossover clinical trials. When applying mixed models to clinical trials data there is also the concern of majntalning the structure, such as strata or sequence, hence the bootstrap samples should be stratified. We present two SAS macros: The first selects bootstrap samples from a data flle with multiple (including unequal) observations per subject. The second macro is an extension of the first macro in that it selects stratified bootstrap samples with varying numbers of records per subject, varying number sample, bootstrap, clinical trial, clinical data, proc step, proc sql, macro definition 8
103 SAS MACROS: A Gentle Introduction for the Fearful Ralph W. Leighton macro variable, proc print, where statement, macro loop, macro definition, proc step, macro facility, variable, %if-%then-%else, call symput, dataset, sas log, insurer, &variable, batch job 10
104 SAS Program Efficiency for Beginners Bruce F. Gilsen dataset, where statement, if-then-else, proc step, subsetting if, conditional do loop, input statement, set statement, character variable, proc append, proc print, flat file, numeric variable, sas function 11
105 SAS, FileMaker Pro, and Microsoft Access for Clinical Study Data Management Karen G. Malley proc step, missing value, clinical trial 9
106 SAS/ASSIST for the New SAS User Jeanne M. Spicer Many experienced SASE> users find SASIASSIST so easy to use that they feel no instruction is necessary. But the inexperienced user who could really use SAS 'ASSISTance' is often stymied at the onset due to a shaky understanding of such concepts as - what is the difference between a flat file and a data set? The new user may get through the tutorial, but then can't get ASSIST to work with their own data. This tutorial gives step-by-step instructions on several ASSIST menu functions. Detailed examples for assigning librefs and file refs in the SETUP menu are given. The user is then guided through the creation of a data set and selected statistical, graphical and report procedures. How and why to use the code generated by ASSIST are particularly emphasized correlation, dataset, filename statement, libname statement, txt, work library, regression analysis, proc step 8
107 SAS/EIS: SPRINTing to the Rescue Jay Jacob Wind,
Claude Griffith,
Glenn Phillips		 This paper describes our adventures installing a multi-featured SAS Executive Infonnation System (BIS) for SPRINT INIERNATIONAL Global Voice Engineering. We cover the capabilities and limitations of SASIEIS and demonstrate graphical examples. SPRINT INTERNATIONAL handles millions of international phone calls each month from countries such as Australia, France, Gennany, New Zealand, and United Kingdom (UK), to more than 200 countries including the U.S. As SPRINT INTERNATIONAL expands, it continuously aims to improve its worldwide selVice. To track and summarize call set-up time, holding, and duration, we installed a powerful Executive Information System (BIS) using SASIEIS for Windows on a Dell Pentiwn desktop computer. The EIS takes advantage of many SASIEIS capabilities. To use those capabilities, however, we bad to overcome numerous BIS limitations macro variable, proc format 7
108 SAS/OR Software - What is It? & What Can It Do? Douglas L. McAllaster This paper gives an overview of the project management and resource alloca tion capabilities of SAS/OR. The paper gives examples of the types of opera tions research (OR) problems that SAS/OR can solve. These problems are found in management of complex projects and when making important decisions via math programming proc step, dataset, sas programmer 4
109 Searching High and Low for Similar Observations Patricia A. Datte call symput, temporary array 2
110 SELECT ITEMS FROM PROC.SQL WHERE ITEMS > BASICS Alan Dickson,
Ray Pass		 proc sql, proc step, select, dataset, group by, order by, by group processing, where statement, create table, raw data 10
111 Setting Domestic Disputes with Designeed Experiments and JMP Lucia A. Ward-Alexander,
Melvin T. Alexander		 One day Mel and Lucia had an argument over the quickest route to drive between their home and a nearby shopping center. Mel suggested that statistics could be used to settle the argument. However, Lucia felt the issue was too complicated to be resolved using statistics because of the numerous factors affecting travel time. Mel countered that argument by pointing out that a statistically designed experiment could be used to settle the issue conclusively. This paper will show how SAS software (in particular JMP) assisted the couple in answering the key questions posed by the argument. An important lesson learned from this exercise was: statistical thinking not only works well in science, business, medicine, and industry, but it can also improve domestic relationships one-way anova, anova, dataset 8
112 Simplifying NDA Programming with PROC SQL Aileen L. Yam The programming of New Drug Application (NDA) Integrated Summary of Safety (ISS) usually in'Dol'Des obtaining patient counts, percentages, and other summary statistics such as mean, standard deviation and range. This paper shows 1zow to obtain all of these results with the SQL procedure. While PROC SQL is often percei'Ded as a data retriwal tool, its unique features allow programmers to write compact codes to obtain data summaries for any application similar to the NDA ISS or the safety summary tables in indi'Didual new drug studies. This paper also shows that sweral DATA or other PROC steps can be reduced to one or two steps with PROC SQL proc sql, adverse events, proc step, proc means, macro call, standard deviation, order by, select, create table, group by, macro variable 7
113 Some Clever Things to Do with User Defined INFORMATS Ronald P. Cody missing value, proc step, proc format, if-then-else, input statement, proc freq 6
114 Spec Edit: A FRAME Based Application For Managing Revision Controlled Spec ifications Richard A. DeVenezia,
Kalman J. Gyimesi		 Specification limits are an important part of any system that uses PROC CAPABILITY or requires range cheCking of important process responses. During the development of a product testing data acquisition and analysis system for Ciba Coming Diagnostics, it became apparent that several specification datasets were required. A flexible, easy to use interface was needed to manage the datasets and maintain an audit trail of changes. SASIAF frames were used to implement the interface. The frame design and implementation are discussed, as well as possible enhancements proc step, audit trail 5
115 SQL Joins - The Long and The Short of It Lewis Church Jr.,
Paul Kent		 This document discusses joining tables with PROC SQL. When joining small tables one can just "go with the defaults" and not worry about performance; these defaults do not always scale well to large tables. The issue is clouded even further when one or more of the tables reside in an external DBMS. Understanding the choices made by the PROC SQL query optimizer helps people decide on a course of action they might embark on to improve the join performance of their queries. This talk presents useful information on the SQL query optimizer, and includes case studies that demonstrate a "cookbook" strategy for improving SQL Join performance. Finally, it discusses two relative recent enhancements - the incorporation of magic-sets [MUMICK] and constraint propagation techniques- to the PROC SQL query optimizer data merging, proc sql 10
116 The Adoption of Computer Interface Technologies in Health Services Delivery Richard Switalski premium, vital signs, service provider, adverse events, insurance claims 9
117 The Choice of SAS Software as the Application Development Tool for Your CANDA Barry R. Cohen pharmaceutical industry, clinical research, dataset, adverse events, clinical data, clinical database, concomitant medications, data review, data science, proc sql 9
118 The Integration of SAS/PH Clinical 2.0 into a CANDA Submission Package Ivan G. Pawson The design and implementation of an effective CANDA submission package depends on the successful integration of software components to support clinical data review . ' report generation, and other elements of the regulatory review process. Essential CANDA functionality includes easy access to any part of a submission via the Table of C.ontents, ability to manipulate data, and the capability to View documents and CRF Images. The user interface should be intuitive, require a mininlUm of user instruction, and be based on existing software conventions, for example, the use of Windows style pull down menus and dialog boxes. A CANDA system under development at Genentech incorporates SASIPH Clinical 2.0 into a comprehensive data and document review package. In addition to providing powerful data review and manipulation capability, PH Clinical 2.0 allows integration of products from our own internal Design, Analysis, and Review Tools (DART) initiative data review, clinical data, data cleaning 3
119 The Use of the SAS/SHARE Product in a Dynamic Manufacturing Environment Hans R. Nagl,
James Robertson		 This paper describes the utilization of SAS/SHARE'lM Software at a telecommunications device manufacturer in the United States. PROC SERVER'lM and PROC OPERATE'lM, as well as the installation of SAS/SHARED' on a UNIX server will be discussed. Keywords: PROC SERVER, PROC OPERATE, Multiple Pr0- cesses, Real Time Iotroduction: Electronics manufacturing companies have complex data storage and analysis requirments which necessitate a robust data base management schema. Electronic devices undergo stringent environmental testing before certifica tion for delivery to the end customer, and typically generate copious amounts of data. These tests, which may be performed by multiple test sets distributed around the shop floor (or possibly in different geographical locations), may need to store data asynchronously over LANs or WANs into centa1ized data base(s). This architecture allows test sets to continue operating without delay due to network problems or downtime on the host data base server proc step, libname statement, test set, error message 4
120 This Old House with SAS William H. Couch When asked what tools to bring to the job site, Bob Vila would probably list such items as a saw, a ruler, and hammer. While SAS/AF software FRAME entry would be an unlikely candidate for inclusion, the object oriented methodologies that it supports are ideally suited for the development of a home construction application that demonstrates the versatility of the SAS System. The graphical user interface (QUI) that is inherent in SAS! AF software FRAME entry, provides a natural environment that increases application integration through support for dynamic data exchange (DOE), dynamic link libraries (DLL), and object linking and embedding (OLE). This application utilizes this QUI to explore selected activities related to the construction of a house. Among the activities included are: cost estimation, project scheduling, site selection, and materials management excel, regression analysis 8
121 Tired of Clincial SAS?: An Application for Drug Safety Jim M. Obritz,
Scott T. LaPorte		 dataset, data science, proc freq 4
122 Training in a World of Cost-Cutting and Downsizing Charles E. Shipp,
Kirk Paul Lafler		 Providing training and user support in a world of cost-cutting and downsizing requires clear odds ratio, quality control 6
123 Understanding Object Oriented Programming and the SAS/AF FRAME Entry Barry R. Cohen The SAS· Institute has added an object oriented application development tool to its product line with its SASI AF FRAME Entry. This tool, because of its object oriented approach, offers many beneficial features for application developers. Nonetheless, the principals of object oriented programming (OOP) are not yet wen understood in the SAS programmer community. Nor does this community fully understand how these principals are delivered within SAS/AF, how much you actually apply these principals to code you write, and how much you just use code that is already developed for you. This paper presents the basic principals of OOP by way of an example. A simple application is described functionally, and then its development is described from both the traditional procedural programming perspective and the new OOP perspective. Comments are then made about the merits of the OOP approach sample, sas base, data science 10
124 Understanding the Collinearity Diagnostics in SAS/INSIGHT and PROC REG Charles B. Hallahan The classical nonnallinear regression model, y = xp + E, comes with a number of assumptions. Some are rather strong, for example that E is a vector of independent and identically distributed normal variates, and various tests for heteroskedasticity, autocorrelation and normality can be performed to check this assumption. Other assumptions, for example that the nxk matrix X has full column rank, are a matter of degree and the near singularity of X and its effect on variance calculations can be detected with a variety of co1linearity measures. The appropriatuess of least squares estimation in the presence of outliers and influential observations can also be assessed. Each potential problem has suggested remedies. For example, transforming data and performing generalized least squares for heteroskedasticity or autocorrelation, robust regression for non-normality or outliers, and ridge regression or principal component regression for co1linearity proc princomp, regression analysis, condition index, proc reg, proc step 4
125 Use of a Date Macro Algorithm to Establish a Moving but Fixed Length Date Range Joseph Pasquale data step, proc print, proc step, macro variable 2
126 Use of a Date Macro at the Beginning of Programs Rather than Find and Replace in Multipl Locations Joseph Pasquale data step, proc print, proc step 1
127 Use of the Call Execute Routing to Pass a List of Variables From a DATA Step to a PROC Step Robert Musterer,
Owen Jiang		 To make a PROC statement generic, the list of input options should be dynamic. Therefore the number of variables passing into is unknown. Usually, the variables are a list of the values from a data set. By using the symput call routine to create macro variables to pass those values, limitations can be encountered in the form of the input variable's length and/or the number of macro variables which must be handled. This problem can be solved if we create a macro variable containing all the options with their list of values from the data set by using the call execute routine. Example: To facilitate project data library integration, we create a project version format library with memtype = catalog and name=formats in the same library as the data. Those formats are used in the project and are copied from different "standard" format libraries. The number of libraries, number of format types, and number of format object names are decided by the specific study project macro variable, proc step, call execute, macro definition 3
128 Using Autocall Macros as a Common Codebase of Data Checks for Clinical Trials: Examples Drawn from the Pharmacetutical Industry Michael E. Tomb Referencing libraries ofre-usable code by way of the SAS Macro Autocall facility is an effective method of minimizing program editing and maintenance. This paper describes a system that FISONS Pharmaceuticals has devised for a common codebase of Clinical Trial data checks and listings. All variable names, arrays, formats, dataset names and pivotal variable values are referenced symbolically. The common code is applied using a two step process. The user may override or accept default values, names and variable typing for all symbolic data cleaning, clinical trial, sample, adverse events, flat file, macro definition, dataset, macro variable, autocall, clinical data, proc step 10
129 Using Base SAS Efficiently Bill MacHose proc step, proc print, cpu time, proc append, decimal, numeric variable 7
130 Using Features of SAS Software to Improve Your Documentation Bill Calvert,
Lynette Swartz		 Specific features of the SAS system are presented in the context of improving the quality and usability of program documentation. Each feature is presented with examples of how it can be used to solve common problems and improve documentation quality, SAS features such as comments, titles, footnotes, and simple macros will be considered in addition to reIated coding practices. The importance of program and system documentation is discussed with emphasis on why it is important for anyone person or work group to establish and maintain quality documentation and how the basic elements of program documentation can be achieved with minimal effort macro variable, sample 6
131 Using Global Macro Variables as a Data Management Tool Cynthia A. Stetz dataset, data step 2
132 Using Indexes for Direct Access to SAS Datasets Judy Loren index, index creation 2
133 Using Object Linking and Embedding, and Open Database Connectivity, With SAS 6.10 and Microsoft Excel 5.0 Robert S. Workman,
Peter Intarapanich,
Gary Drucker		 excel, proc sql, sample 3
134 Using PROC EXPAND with Time Series Data Andrew H. Karp proc step, proc freq, dataset, time series analysis, missing value, sysdate9, datetime variable 4
135 Using PROC GPRINT and Enhancing The Printed Output Edmond Williams,
Shi-Tao Yeh		 proc step, txt, sample, proc greplay 5
136 Using SAS To Compute Variances For Stratified Samples Lawrence Helbers Standard SAS procedures for computing sampling variability for numerical variables do not use the appropriate error model in the case of stratified samples. The standard error estimate reported by SAS can be either larger or smaller than the desired estimate depending on the specific situation. Of course, SAS's great flexibility allows analysts to compute the desired statistic, but this requires multiple steps and programming. Below a SAS macro is discussed which simplifies this . computation sample, proc calis, standard deviation, confidence interval, proc means, proc step, macro variable, sample size, macro definition 5
137 Using SAS with SYBASE For Database Management Yusuf M. Mohanedshah,
Carolyn D. Williams		 proc sql, proc step, numeric variable, dataset, character variable, proc sort, if-then-else, sql server, proc contents 8
138 Using SAS/AF and SAS FSP for Development of OO Based Enterprise Applications Carl LaChapelle 6
139 Using SAS/GRAPH For Data Visualization Frank J. Tedesco data step, proc greplay 2
140 Using the SAS System to Create an Interactive Reporting System Using a Host Specific DBMS Philip Hamm,
Rodney Meeker		 proc sql, macro variable, proc step, dataset, where statement, sas base, data step, excel, select 8
141 Using the WHERE Statement Bob Virgile proc step, dataset, proc freq, where statement, sample 5
142 Utilizing SAS To Creat A Patient's Liver Enzyme Profile Erik S. Larsen proc step, macro variable, proc gplot, proc greplay 6
143 VSAM Files Made (Almost) Simple; SAS System Approach Mary A. McDonald data step, length statement 2
144 What We Learned At NESUG: An Application of Lesson Learned at NESUG '91, '92, '93 & '94 Charles A. DePascale,
Patricia Cole		 missing value, input statement, proc step, if-then-else, put statement 4
145 WHATITDO: Documentation Automatically Produced By Your SAS Programs Karen Crandall proc step, proc freq 3
146 When Chi-Square Does Not Hold: How to Analyze Tables With Low Cell Frequencies With SAS Lothar Tremmel Detennining if two proportions are significantly different is a common problem in statistics. For low sample sizes, the common chi-square tests may give the wrong answer, because normal approximation is no longer valid. Exact tests such as Fisher's exact tests need to be used instead. The randomization model for such a test is developed, and it is demonstrated how to obtain this test with SAS procedures and functions. Oftentimes, one has a set of pairs of proportions, such as the proportion of responders under Active treatment vs. the proportion under Placebo for males, and the corresponding proportions for females. 'Gender' is the stratification variable, and 'male' and 'female' are the 'strata'. It is often desirable not to pool these data into a single 2x2 table, but to estimate the treatment effect in each stratum separately and to combine the estimates. An exact test is developed, and it is shown how to calculate the test with the new SAS procedure MUL 1TEST p-value, sample, exact test, proc freq, proc step, hypergeometric distribution, sample size, binomial distribution, chi-square test, chi-square distribution, treatment effect, randomization, normal distribution 9
147 Working with Classes in Version 6.11 of SAS/AF FRAME Entries Daniel Gronell dataset, sample 8
148 Working with SAS Date and Time Functions Andrew H. Karp sysdate9, date variable, datetime variable, datepart, time variable, numeric variable, dataset, raw data, proc freq, time series analysis, if-then-else, sas function, subsetting if, proc step 5

Back to Conf go 1996 go 1994

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.