HOME QUICK SEARCH AUTHORS KEYWORDS HELP BUGS

WUSS 1996 Papers - Index

68 papers. Source: WUSS (Western Users of SAS Software) conference proceedings.

Back to Conf
go 1997
go 1995
# Title
×		 Authors
×		 Abstract
×		 Keywords
×		 Extras Pages
1 "Rank Ordering" Means Diana D. Suhr Presenting results from your data analysis can be a challenging task. Very often information is not arranged in the desired form. As an example, a data analysis for a needs assessment included a PROC MEANS procedure on twenty-five questions. Instead of takiJig the time to rearrange the results in order from the largest mean value (most important) to the smallest mean value (least important), the data was manipulated and a rank ordered means table was created. You will be shown a strategy to ''rank order" means using PROC MEANS, PROC SORT, PROC FORMAT, output statements, and array statements. proc step, proc format, proc sort, proc print, output delivery system, proc means source 5
2 %BYVARS: Determining the Order of Variable Names in a Macro Parameter Standish Sibley This macro tool was written for an elaborate computer model which requires very flexible possibilities for the analysis variables. A user-supplied parameter lets the model know how detailed the analysis is to be by specifying a list of variables used for summarizing and/or sorting. The macro places the names of the analysis variables into global macro variables, and places the number of variable names encountered into a macro variable as well. In this way, the system can use the FIRST. and LAST. conventions for summarization and break points without actually being told what variable(s) are the lowest in precedence, next-to-lowest, etc. class variable, proc step, proc means, macro variable source 3
3 A Process for Simultaneous Porting of Applications to Many Platforms Richard D. Langston As part of the development process for the SAs* File Conbibution Repository, the author found It necessary to streamline the process of porting SASITOOU<IT applications to many different platforms. This paper describes the mechanism implemented to allow for simultaneous porting of such applications to all these platforms. sample, proc step source 2
4 Breaking Down the Barriers Using ODBC to Share Data Between SAS and Other Applications Andrew J. L. Cary SAS and ODBC provides a mechanism for moving data between SAS and Microsoft Windows applications. SAS can be the 'client' and receive data from ODBC compliant 'server' applications such as Microsoft EXCEL, Microsoft Access, and Oracle. SAS can also be a server and allow ODBC compliant applications to access native SAS data set sand views. This would allow a user to use a product such as Crystal Reports as a report writer or MS-EXCEL as a spreadsheet data with SAS data sets. This paper discusses when and how to use ODBC and SAS. What is ODBC? ODBC (Open Data Base Connectivity) is a vendor independent data translation and sharing protocol. Originally a Microsoft specification for the MS Windows platform, it has since been adopted by most major vendors. It provides a common data description language. This allows one application to request data from another application without knowing the internal storage for that application. proc sql, proc step, dataset, excel, libname statement, odbc connection source 10
5 Building a Simple SAS Macro to Generate SQL Instructions in Frequently Used DB2 Tables Monique Bryher Writing SQL instructions to extract data stored in DB2 can be made easier through the use of simple coding techniques that utilize SAS macros and the SAS Autocall Facility. Symbolic variables are replaced with values for common SQL expressions in frequently-accessed tables; this has the advantage of minimizing coding time, reducing syntactical errors, affording multiple individuals access to code stored in a "tool kit" of macros that have been tested, debugged, and which can be documented. As will be shown in this paper, these techniques can also be employed to generate large lists in a WHERE predicate that would otherwise be impractical to enter manually. proc sql, create table, macro put, proc step, autocall macro, macro variable, order by, macro definition source 5
6 Can SAS Programs on PCs and Minis Finally Compete with Mainframe SAS Programs Robert O. Bernstein A set of more than 20 large mainframe SAS programs using ten different data sets that totaled nine million automobile insurance records, cost the Department of Insurance (CDI) $94,000 in mainframe charges in just one month. CDI purchased two new systems to replace the expensive mainframe system. These systems, one Unix Mini and one PC, ran one of the largest of the 20 mainframe SAS programs using the largest data set, three million records, to decide if they could replace the mainframe system. elapsed time, proc step, cpu time, proc univariate, if-then-else, data step, proc sort, proc means, proc rank source 6
7 ClinAccess: An Integrated Client/Server Approach to Clinical Data Management and Regulatory Approval Martin J. Rosenberg Motivated by the desire to better manage an ever increasing volume of information and to reduce the length of time required to introduce new drugs around the world, the pharmaceutical industry and regulatory agencies such as the U.S. Food and Drug Administration have sought to facilitate the drug development and approval processes through innovative uses of computer technology. One such effort that has received great attention is the computer assisted NDA or CANDA and its cousin for biologicals the computer assisted PLA or CAPLA. The first CANDA's were created with much effort, and at great expense, after the paper NDA's had been filed. From these early experiments, it became evident that if CANDA's were to ever become routinely used throughout the industry by both large and small companies, their development would need to be simplified and to begin earlier in tlie drug development process. clinical trial, clinical data, crf, data review, data governance, clinical data management, audit trail, clinical research, pharmaceutical industry, key, dataset, data cleaning source 9
8 ClinOS: A Commercially Available Set of Tools for SAS Programmers Lindley Frahm proc step, proc sort, data science, clinical trial, macro variable, clinical data source 8
9 Comparison of Different SAS Modules for Effective Data Analyses George C. J. Fernandez The SAS system has a variety of powerful and user-friendly modules for performing simple and advanced statistical analysis. In this tutorial, data analyses using examples in ANOVA, multiple linear regression and multivariate techniques will be demonstrated using SAS modules, SAS/ASSIST, SAS/ LAB, and SASIINSIGHT. In addition, integrated data analysis using SAS macros will be tom pared with these dynamic SAS data analyses modules. data science, regression analysis, anova, correlation, sas programming, proc freq, histogram, box plot, macro definition source 2
10 Contrasts, Cell Means Model, and Missing Cells Joseph Campbell Complex contrast matrices for many factors can be easily and intuitively developed, using simple contrast matrices for single factors. Using SAS/MACROS and SAS/IML, an algorithm using recursive Kronecker products can make painfully large contrast matrices easy to create and use. An application of the algorithm will be shown in recovering estimability of missing celJs, using the cell means model approach. macro definition, data science source 4
11 Cross Platform Generation of Reports from Oracle Using SAS on MVS, Macintosh and AIX Marvyn M. Lindsey The objective of this paper is to discuss the implementation of a system to generate reports from an Oracle database using the unique features and functionality of SAS on each of three different computing platfonns: MVS, Macintosh and AIX. This paper will discuss how the CSU Chancellor's Office Corporate Information Systems team developed a system to generate over 300 annual reports from an Oracle? database using SAS/MACINTOSH. These reports had been previously generated via COBOL from flat files on MVS. The methodology that was developed makes use of SAS/CONNECT and Base SAS on all three platforms, SAS/ACCESS for Oracle and PROC REPORT on AIX, and SASIEIS and FR~ on the Macintosh. proc step, proc report, proc printto, batch job, sample, flat file, quality assurance, sas programming, elapsed time, macro variable source 6
12 Data Know Thyself: A Macro to Tell You Everything You Need to Know About Data Set Characteristics Karsten M. Self Are you faced with the challenge of interpreting numerous internal and external 'mysteiy' datasets? The VV macro illustrates the use of the SAS System's metadata resources and code-generating capabilities to develop a solution which automatically and intelligently generates useful descriptive information about a SAS data set in a concise, highly informative, format VV provides information derived from the SQL, FREQ, UNN ARIATE, CONTENTS, and PRINT procedures. This paper describes the use of and development methodology ofVV. call execute, proc freq, proc sql, proc step, character variable, macro facility, numeric variable source 5
13 Data Warehouse: A Telecommunications Business Solution Demetris Papaiacovou,
Larry D. Bramblett,
June Burgess		 The telecommunications industry is rapidly becoming the largest user of high-volume information systems. This is especially true in the United States, where telecommunications, Internet, video and audio services all fall under the larger umbrella of the information superhighway. Our project team is adept at providing customized business solutions to telecommunications managers by creating specialized information systems to meet their individual needs. We design personalized systems which transcend the traditional boundaries of data warehousing systems by gathering and storing important data, analyzing and transforming the data, and then summarizing and reordering it in accordance with the demands of the user. This provides a dynamic and efficient front end presentation and reporting tools that transform the users from reactive recipients of hard copy reports to competitive executives poised to make efficient, well informed decisions. This paper will discuss how the power and versatility of the SAS system were used to provide data warehousing business solutions to Pacific Bell, one of the largest telecommunication companies in the United States. data warehouse, data cleaning source 5
14 Data Warehousing with the SAS System Terry L. Lewis Few topics have taken hold of the computer industry the way that data warehousing has over the past few years. The SAS System has long been used to provide data warehouse functionality. The SAS System provided this functionality even before the term data warehouse was coined. This paper will cover three topics: why has data warehousing become such an important issue; an explanation of data warehouse architecture and how existing tools in the SAS System may be employed in the design and implementation of that architecture; a description of some new tools from SAS Institute that will facilitate faster and easier data warehouse implementations. Data warehousing consists of a series of data refinements and reformats in an effort to move data from a operational design to a design that more closely reflects decision support applications and a business model. The paper will explore data warehousing concepts in general and then will drill down into how the SAS System can be used to make the concepts a reality. data warehouse, proc step, sas base, dimension table, dataset source 5
15 Date Variables in the Year 2000 Amber Elam sysdate9, dataset source 2
16 Design, Analysis and Review Tools for Clinical Drug Development Using SAS Object-Oriented Products David H. Christiansen clinical research, clinical trial source 8
17 Designing and Implementing SAS Programming Standards: Examples and Suggestions Miriam G. Cisternas,
Marcia Stoddard		 Have you ever tried to debug or revise a coworker's program? Replicate results produced several years ago by a stranger ... possibly yourself in a previous life? When working in an environment with other programmers, there are several reasons why it is important to have standards for writing and documenting SAS programs and data sets. First, standards facilitate transferring projects from one programmer to another and having multiple programmers work together on the same project. This flexibility is crucial to ha'.(ing quick and accurate turnaround on projects in any environment where tight deadlines are the norm. Second, standards aid in the debugging process, both for a programmer working alone and in a group trouble-shooting environment. Finally, the process of developing and implementing programming standards helps in the training of new hires. This paper is divided into two parts. We first outline areas for which we think standards should be developed and provide examples. quality control, sas output source 5
18 Developing Applications with the SAS System Emphasis on the SQL Query Window Julianne Staub Perry,
Henrietta Cummings		 The desktop has become an Important arena for the generation of key applications in many enterprises. It is also the playground for the most innovative technologies for applications available In the industry. This presentation will demonstrate the methodologies that SAS software uses that allow you to take advantage of the graphical user Interfaces (GU I) and the features of object oriented (00) rapid applications development. You wiH be introduced to two development tools, SAS/AF and SASJEIS software. Included in the discussion will be a comparison of these two tools and how you can use each to develop applcations either Independently or in combination with one another. Included will be a discussion of an application that utilizes the SQL Query Window to provide client server query and reporting capabilities. Also included is a behind the scenes look at SCL driven SASIAF FRAME Entries. source 10
19 Dressing for Success with SAS Software Functions Faith Reneé Sloan What are SAS functions? A technical definition is that a SAS function accepts none or many parameters and upon execution, returns a value from a computation or system manipulation. Practically speaking, SAS functions are tools which the SAS Institute have developed in order to make our programming lives easier. There's a vast array of functions provided with the SAS System Software. Why are functions useful? These tools perform arithmetic, string processing, mathematical, statistical, financial, and other functions which eliminate the need for programmers to write extensive code to get the job done. SAS has done the dirty work for us. In this tutorial, the practical use of some of the many SAS functions will be demonstrated. Strips fupqtiogs XRDBX(string,search) - searches 'string' for 'search' and returns the position of the first occurrence of the *search' parameter. missing value, character variable source 5
20 Engineering Efficeincy into SAS Applications Kirk Paul Lafler With the ever increasing popularity ofSAS Software among users and programmers, guidelines for its efficient use has widespread importance. This paper presents a collection of techniques for gaining efficiency when using SAS Software as an applications development tool. Areas receiving special consideration include program execution, I/0, disk space, memory, and program maintenance. Numerous practical and proven techniques will be examined using sample code to illustrate why some methods are better than others. proc sql, cpu time, drop statement, if-then-else source 6
21 Exploratory Variable Clustering for Integrating Analyses David J. Pasta,
Lori Potter		 In a cross sectional study comparing treatments for pain, we demostrate how PROC V ARCLUS was used to strengthen and simplify results and discover the underlying structure in the data. Analysis was begun using standard estimation procedures (PROC GLM) with a large number of outcome measures. This paper traces the steps involved in developing composite scores, and discusses how to apply clustering techniques to data that has both single item, scale, and subscale scores. A proposed proc princomp, r-squared, proc step, proc freq, correlation source 5
22 GUI to CRSP and COMPUSTAT Data Peter C. C. Luan CRSP (Center for Research in Security Prices) data and Standard & Poor's COMPUSTAT data are probably the most often used data for instructional and research purposes in many universities and financial institutions. This paper shares the design, describes the functionality of two graphical user interfaces (GUis), and details some of the experience of developing these two interfaces. proc step, character variable source 4
23 How I Developed User Friendly Applications in a Batch Environment and Lived to Write About It Curtis A. Smith All those nifty SAS GUI modules really make a user's job easy to access and customize automated SAS applications. Selecting options, screening criteria, and other things can all be done with a click of a mouse or a few keystrokes. Unfortunately, not all users and developers have the fortune of having access to the GUI modules or even the interactive environment. Problems like tape access, large data warehouses, and processing cost issues can eliminate the interactive environment from consideration. Further, cheap executives who won't purchase SAS/AF and other GUI modules can keep a developer in the dark ages. Yet, with a few well-thought techniques, the developer can effectively automate SAS applications in a batch environment. libname statement, proc step, macro variable, proc report, batch job, dataset, macro loop source 7
24 How Symbolic Variables Can Reduce Code in a Graphics Environment Monique Bryher This paper discusses how to generate a series of 4- way bar charts that are replayed in SAS/GRAPIFL when the list. of possible values assigned to the independent variable is changeable. Macro language coding removes the necessity of maintaining a hard coded list that might have to be modified during the reporting period, e.g., monthly, by having SAS assign each value of the independent variable to a symbolic variable, which is then resolved during the production of the graph. This eliminates the need for writing code to create each graph, resulting in a program that is shorter and easier to maintain. Last, examples of code will be given which illustrate the power of using double and triple ampersands derived from data variables to supply information required by graphics language expressions, such as. names, labels, titles, and footnotes. proc step, macro variable, call symput, bar chart, catalog entry, macro definition source 6
25 How to Get People to Work REAL Hard for No Money: Managing Volunteer Organizations Claire Castelle source 3
26 In-House SAS Training Martin Haase proc step, data step, proc sort, proc print, proc freq, raw data, proc sql source 6
27 Introduction to SAS/ACCESS Software Andy Wilcox source 2
28 Loading SAS Data in Informix: An Integrated Approach M. Kumar Sagar In a typical clinical trial, especially large and/or multicenter ones, there are many sources of data, including electronic data transfers from sites, central labs and CROs. While the data may come in many formats, the question arises: how to load these data in to appropriate databases. In this context, we have taken a holistic approach whereby SAS is used as the platform of choice to deal with these data transfers: SAS transport datasets coming in are converted to host SAS datasets, and required data transformations and manipulations are applied. Once the data is in the desired format, an output file is written and then, using SAS/ACCESS for lnformix, an lnformix table is created and populated with the data from the SAS datasets. In this process, UNIX utilities are also used as an umbrella, to perform various tasks such as input verification, error checking, invoking SAS programs, cleaning the data, and to archive the associated files. This dataset, proc step, data cleaning source 5
29 Logistic Regression: From SAS Coding to Statistical Interpretation Elizabeth Li Interpretation of SAS PROC LOGISTIC outputs can be difficult. SAS manuals primarily focus on models and SAS codes while statistics books emphasize hypotheses, models and interpretations. This paper links hypotheses, models, SAS codes and interpretations. It explains the codes and interpretations of results, enabling more accurate use of PROC logistic distribution, odds ratio, predictor variable, proc logistic, regression analysis, proc step, likelihood function, response variable, maximum likelihood estimation, null hypothesis, sas output, logistic regression source 9
30 Mainframe Directions Gary Burchett,
David Quattlebaum		 While SAS development interest has seemed to be focused on desktop platforms, mainframe development staff have been quietly keeping Version 6 of the SAS System up to date with their respective Operating Systems, adding new functionality, and planning for the future. This paper discusses some of the ongoing work, portions of which have been delivered along with maintenance to the 6.08 SAS System, and some of our thinking about the future direction of the SAS System on the mainframe systems. source 5
31 Measuring Performance Between PROC SQL and the Merge Approach Martin Haase proc sql, proc step, cpu time, proc sort, elapsed time, create table, benchmarking, insurance claims source 5
32 Minimizing CPU Time of SAS Software Applications and Getting Rewarded For It Daniel J. Leprince Optimizing SAS applications often results in substantial savings of computer resources which can translate into significant financial gains for your institution. The successive steps taken to improve the efficiency of a SAS application in an IBM MVS environment will be presented. Topics covered will range from obtaining performance statistics, redesigning data flow, programming tips, and relating the savings to your administration. cpu time, elapsed time, fullstimer, proc print, proc step, proc contents source 5
33 Multi-dimensional Data Model Extensions to Data Warehouses John McIntyre source 8
34 Multicollinearity in Regression Models: Detection and Remedial Measures Using the SAS System George C. J. Fernandez In multiple linear regression models few problems arise when serious multicollinearity (two or more predictor variables have large correlations) is present The consequences of multicollinearity are rounding errors in the estimation of regression coefficients, variance inflation of parameter estimates, unstable regression coefficients, and contradicting significance tests between the overall model and individual parameter tests. Multicollinearity can be detected graphically by the overlaid partial-regression and partial-residual plots, VIF and eigen system analysis and condition indices. The remedial measures for multicollinearity include variable selection and redefinilion, ridge regression, and incomplete principal component regression. The features in SAS systems for detecting and correcting multicollinearity are discussed here. regression analysis, predictor variable, proc princomp, ridge regression, multiple regression, proc reg, correlation, condition index, macro definition, sample source 9
35 Object, Object, More Objects K. Deva Kumar,
Tammy L. Gagliano,
Patti Brideson		 The class library in Release 6.11 of SAS/AF software has expanded to include many new visual and non-visual classes. This paper takes a closer look at some of the visual classes and what features they will add to your FRAME applications. dataset, catalog entry, data cleaning source 8
36 P's Porridge: What's In Those Brown Manuals? Jack Hamilton source 5
37 Plots With Generalized Confidence Intervals Using SAS/GRAPH Linda C. Whitehand,
Bruce E. Mackey		 Plotting of group means and confidence intervals is a popular method with the statistical community for summarizing data. The technique provides more information than multiple comparison procedures and is easier to comprehend than tables of the same statistics. Although SASTM PROCs GPLOT and GCHART provide ways of plotting means and confidelice intervals, they are quite restrictive for many types of complex analyses. For example, they do not easily plot asymmetric confidence intervals, which occur for binomially distributed data. Asymmetry also arises with data requiring statistical analysis in a transformed scale, but for which presentation of means and confidence intervals is desired in the original scale. Even with symmetric cases, derivations may be complex and/or confidence intervals may be unequal among groups, such as with those produced using PROC MIXED. confidence interval, proc step, proc gplot, data science, proc genmod source 5
38 Preliminary Data Mining for Subgroups in a Hospital Inpatient Population Anthony M. Dymond A business plan to manage a group or population depends on knowing if the group is homogeneous or composed of several subgroups. This paper discusses data mining and describes some practical considerations for preliminary data exploration in a hospital inpatient population of 5760 unique patients. A variety of visual and analytic examinations of selected variables suggest that three subpopulations may exist in this data. data mining, discriminant analysis, outlier source 6
39 PROC SQL - A SAS Programmer's Perspective Deborah L. Skinner proc sql, proc step, proc print, group by, sas programming, order by, proc sort, where statement, insurer source 7
40 PROC SQL as an Alternative to Multiple SORT/MERGE Statements Robert Bachler Frequently, pulling data from SASe data warehouses requires accessing several different data sets. The SQL procedure foWld in BASE SAS is a powerful tool that is ideal for dealing with these situations. A single SQL procedure can replace groups of SORT procedures and DATA STEP MERGE statements. Not only is · SQL code more concise, but is often easier to read and more efficient This paper will show examples where the SQL procedure can replace multiple SORT/MERGE statements. Benchmark results will also be shown from examples using different numbers of observations, and indexed and non-indexed data sets. proc sql, proc step, proc sort, cpu time, insurance claims source 3
41 Repeated Measures Analysis with the SAS System Kathleen Kierman This paper will present several statistical methods for the analysis of repeated measures data. The traditional uni variate and multivariate methods of PROC GLM will be contrasted to the mixed-model approach of fitting a class of covaiiance structures in PROC MIXED. A regression ap proach to repeated measures data will also be discussed. proc step, covariance, repeated measures, anova, proc mixed, regression analysis, proc genmod, estimate, fixed effects, proc glm, longitudinal data, covariance structure, degrees of freedom, likelihood function, compound symmetry, f-test, p-value source 10
42 Resolving and Using &&VAR&I Macro Variables Arthur L. Carpenter One of the most important aspects of the Macro Language within the SASe System is its ability to create code dynamically. This allows the developer to write code that will do a series of operations without knowing exactly what those operations will be at execution time. This presentation will show how double ampersand macro variables are resolved and how compound macro variables can be used to create dynamic code. Examples include using a list in a flat file to create a set of macro variables that control a series of events. macro variable, flat file, proc step, by group processing, call symput, macro loop, %if-%then-%else, macro definition source 4
43 SAS Automatic Forecasting Methods in Practice Margery H. Connor Chevron has been using the SAS multivariate statespace model to forecast demand for unleaded, mid-grade, and super unleaded gasoline at the terminal level. This paper reports on an investigation into the SASIETS software's other automatic forecasting methods: stepwise autoregressive, exponential smoothing, Winters, and Addwinters. We compare the performance of these methods with the statespace model, and we discuss lessons learned on multivariate time series, seasonality, and the training of users. proc step, proc esm, proc forecast source 2
44 SAS Macros and Fraud Detection in the Department of Defense Wayne F. Woo,
Joyce Hamza		 Operation Mongoose began in June 1994 as a new Department of Defense (DoD) Initiative to detect and prevent fraud In DoD financial systems. Within the Vendor Pay system, one of the fraud indicators was a single vendor with multiple payee addresses or conversely, multiple vendors with the same payee address. Payees would occasionally submit vouchers with slight variations of their "pay-to" names and/or addresses. The Defense Manpower Data Center (DMDC) created and maintained a database of unique payee names and unique payee addresses, and each name and address was assigned a unique system code. SAS macros were created to "clean and standardize" submitted data before matching against the databases to obtain system codes. This paper/poster describes how SAS is an important part of this project. data step, outlier source 5
45 SAS PROC MIXED to Fit Growth Curve Models: Application to a Longitudinal Group Randomization Study Biao Wm. Lu,
Sandra Wilson,
David Latini,
David Dayton		 A brief account is given for using PROC MIXED to fit linear growth curve models to repeated measure data from community clinical trials. Issues about modeling the clinic effects when randomization is done at the clinic level, interpreting the random effects coefficients, model assessment, and modeling the time-dependent covariates are discussed. Real examples will be used to illustrate the approach. proc mixed, randomization, proc step, fixed effects, clinical endpoint, random intercept, restricted maximum likelihood, clinical trial, likelihood function source 6
46 SAS System Programming Tips and Efficiencies: A Tutorial Andrew H. Karp dataset, data step, if-then-else, proc step, raw data, output, where statement, sas programming source 8
47 SAS Tools and Internet Charles E. Shipp,
Kirk Paul Lafler,
Sunil Kumar Gupta		 The Internet continues to grow rapidly in ease-of use, value of content, and popularity- many new users, both companies and individuals, join every day. Part of its magical attraction is its wide diversity of information and modes of search and display. From e-mail and discussion groups, to downloading text and programs, a wealth of valuable and timely information awaits you on the Internet. Jump in! SAS Institute and professional SAS programmers and consultants have a strong presence and tradition on the Internet, and new innovations are planned and being implemented. source 4
48 Survival Analysis with Staggered Entry Times and a Time-dependent Covariate Alison J. Canchola I demonstrate a Kaplan-Meier survival analysis with (1) staggered entry times (left censoring) and (2) a time-dependent covariate. (1) Each subject has an entry time {the unique time of enrollment into the study) and an exit time (follow-up time until censoring or the outcome event occurs). A counter Is used to count the number of subjects at risk (under follow-up) at each distinct event time. The survival probability is then calculated at each event as is done in PROC LIFETEST. (2} A time-dependent covariate, which equals 0 until a predictor event occurs and 1 thereafter, is incorporated into the survival plot, yielding two survival curves. Any subject whose covariate value does not change appears in only one curve. However, a subject whose value changes from 0 to 1 leaves the first curve and enters the second at the time of the change. time-dependent covariate, proc step, left censoring, survival analysis, proc phreg source 5
49 System Development Life Cycle for SAS Applications Development Incorporating a Procedure for As-Built Validation of In-Use Products Wayne Lanier,
Deb Gilmore		 SAS is increasingly being used as an applications development language. Originally an analysis lan guage, SAS tradition Jacks the System Development Life Cycle (SDLC) procedures found in languages with a history of software publication. Of concern is validation for SAS-developed systems supporting clinical studies in application for FDA drug licenses. Resource Biometrics, Inc. (RBI) is a Contract Re search Organization (CRO) developing SAS-based software for Data Management and Data Analysis in the biotechnology and pharmaceutical industries. We are now publishing these software products for sale. It is necessary to validate these systems as part of current Good Clinical Practice (cGCP.) Initially we defined a simple RBI System Develop ment Life Cycle for new-product development which sets out the tasks for development, validation, re lease, and maintenance of software products. data cleaning, clinical trial source 6
50 The Effect of Poison Pills on Firm Risk: An Application of Options Pricing Theory Thomas A. Turk,
Jeremy Goh		 Many researchers have attacked poison pill securities on the grounds that they insulate managers from the efficiency enhancing effects of an active takeover market. Others have argued that they serve to counter the risk-reducing pressures of an active takeover market. This study employs the Black and Scholes Option Pricing Model to determine whether or not adoption of poison pill securities increases investor perceptions of firm risk. The results provide evidence that the Option Implied Standard Deviations of common stock returns increase significantly on the poison pill adoption date, on average. Furthermore, the implied standard deviations remained significantly above pre-adoption levels for several days after the poison pill adoption, suggesting that the perceived increase in firm risk is permanent. standard deviation, in= option, sample source 9
51 The Incredible Shrinking Report Diana D. Suhr Presenting results from a data analysis can be a d.ifficuh task. We are constantly challenged to balance the amount of information presented with simplicity. This poster session will provide a specific strategy for reporting information which can be used as a prototype to create your own reports. The presentation shows how to create a single page table for twenty-five questions, using a needs assessment survey as an example. The single page table presents information more simply and concisely than twenty five separate frequency tables. Program code shown in the presentation includes array statements, output statements, PROC TABULATE, and PROC proc step, proc tabulate, proc freq, array source 5
52 The Intraclass Correlation: What Is It and Why Do We Care? David J. Pasta The intraclass correlation coefficient (ICC) is a measure of reproducibility that should be used in many common situations where the ordinary (Pearson product-moment) correlation coefficient is used inappropriately. The ICC is defined and its history and motivation described. Alternative methods of calculating the ICC are presented in detail. The ICC is related to other measures of agreement, such as the concordance correlation coefficient. correlation, intraclass correlation, proc mixed, anova, standard deviation source 6
53 Tips and Techniques for Updating a Repository Data Warehouse Ty Lee A repository data warehouse is popular for statistical analysis and reporting. It is often necessary to update a data warehouse. This paper will discuss tips and techniques for data updates in a SAS repository data warehouse. Specifically, the paper will discuss alternatives for deleting, inserting and changing· records, combining tables, and validating results. The paper will also demonstrate how to use metadata and macro facility to automate global data updates in a SAS data warehouse. To make it more interesting, we will employ the case study approach to explore the following topics: * The Data Warehouse Environment Data Update in a Nutshell A Data Driven Approach for Changes Efficiency Considerations PROC SQL vs. DATA STEP Validation Issues Where to fix? - SAS Data Dictionary Data Warehouse vs. Database Both database and data warehouse are collections of data. They differ in their purposes and structures. proc step, data warehouse, proc compare, audit trail, proc sql, data dictionary, merge statement, proc sort, data cleaning, macro definition source 6
54 Tips, Tricks and Traps Using SAS on the Mainframe Ron Ridley This tutorial describes different techniques for selecting and manipulating data. Software discussed will be SAS BASICS, IBM MVS JCL, DYLAKOR's DYL-280 and SyncSort Incorporated's SyncSort MVS release 3.6. Mainframe costs for accessing data can be high. Therefore, DYL-280 or SyncSort can be used on large tape inputs for calling in specific records and to accept only certain fields (variables) needed for further processing by SAS BASICS. Multiple files can be called into a program using different techniques with the Job Control Language (JCL). You can call in several direct access sequential disk (DASD) files using a single input statement and multiple Data Definitions (DD) with data set names (DSN). Several tape inputs can be called in sequence by using a UNIT=AFF statement on subsequent files. SAS BASICS has many statements for selecting data such as WHERE, WHEREIN, IFIN, and MERGE. Applications designed in this modular fashion can save mainframe processing costs and allow a tracking system for monitoring the changes in your data as the finished product is being developed. These programs can be used over by modifying certain parameters and source code to accomplish different tasks. Using these different techniques makes it possible to call multiple input files, shrink data, and select/summarize/output data. mis, dataset source 2
55 Using Dynamic Data Exchange with Microsoft Word Jodie M. Gilmore Dynamic Data Exchange (DDE) enables you to share data between the SAS System and many other Windows applications, including Microsoft Word. In this paper, you'llleam how to prepare your Word documents so that the SAS System can access them, and how to write SAS code that reads data from and writes data to those Word documents. Also, you 1lleam how to use DDE to send commands to Word documents from your SAS session. filename statement, put statement, length statement, macro facility source 6
56 Using PROC EXPAND with Time Series Data Andrew H. Karp proc step, proc freq, dataset, time series analysis, missing value, sysdate9, datetime variable source 4
57 Using PROC REPORT to Summarize Clinical Safety Data Justina M. Flavin By using the summary and display features available In PROC REPORT, the amount of code needed to generate output can be reduced and often, a more informative table can be produced. This paper wDI illustrate the use of the REPORT procedure in creating both detail and summary reports of clinical safety data such as adverse events, laboratory results, and concomitant medications. proc step, proc report, adverse events, proc means source 5
58 Using SAS to Create WWW Pages Tim Berryhill The paper covers how we use text tables and SAS to maintain a portion of our web server, http://www.pge.com/customer_services/business/tariffs. Topics include SAS coding, business practices and relevant HTML. txt, data step source 4
59 Using SAS to Process the Data on Tape Ying Zhou Southern California Association of Governments (SCAG) is a designated census data center for most of the Southern California. Therefore the data files SCAG deal with are usually very large. Tape is the preferred medium for storing large data files for processing. SAS has the ability to process data on tap~ directly. At SCAG, SAS is used on the IBM RISC6000 UNIX platform. This paper will detail several SAS programming techniques to process data on tape within the UNIX environment. filename statement, proc copy source 3
60 Using SAS Views for Data Resource Visualization Steven A. Wilson SAS Views are extremely powerful tools for managing data files with the SAS System. SAS views are compiled SAS statements that define or describe data that are physically stored somewhere else. The physical data referenced by a SAS view may be a SAS data set or some other file structure that is accessible via SAS. Once created these views are used like any regular SAS data set. In an environment with many different types of frequently accessed data sources, it can be useful to have a shared set of SAS views that are used by a large number of personnel. These views can be used to examine the many different data sources in a similar format. using consistent SAS variable names, labels, and formats. After a brief discussion of the pros and cons of using SAS views, this paper will discuss a method of using SAS views in conjunction with SAS/ AF extended tables for displaying the contents of a data source. The emphasis of this paper will not be SAS/ AF development. proc sql, view, proc step, sql pass-through, dataset, create view, proc print, variable source 10
61 Using SAS/AF Software for Data Warehouse Applications: From Raw Data to Reports Joseph Campbell,
Robert Bachler		 SAs is no longer a language only for technical users. Often, non-technical users are required to report on, summarize, and sometimes maintain data. Given the popularity and ease of windowing environments, a point and click application that will allow non-technical users to perform these tasks would be ideal. This would allow more technical SAS users to concentrate on more difficult problems - Let's face it, we like to write the programs, not run them. This paper will show an example of an application written in SAS/ AF software that controls all aspects of a Data Warehouse. Raw text files are read into permanent data sets, formats can be checked and updated to deal with new aspects of data, and reports are generated as an end product. Key concepts: entiy values, macros and macro variables, formats control data sets, display manager commands, external files, submit blocks. txt, proc step, macro definition source 6
62 Using Survey Data: Tips to Pick Up Speed on the Road to Analysis Cynthia L. Williamson,
Suzanne D. Kreutzer		 In this paper, we offer tips to help SAS programmers get to know their survey data, thereby ensuring that the road to analysis is smooth. We will address techniques to prepare for a survey, track questionnaire returns, ensure accurate data input, check for data quality, and create a data analysis file. proc step, sysrc, sample, proc compare, proc sort, dataset, proc freq, proc print, outlier, macro definition, missing value source 6
63 Using the GEE v2.02 SAS Macro to Adjust for Clustering of Respondents Within Interviewer in a Methodological National Probability Survey/Study Jesse Albert Canchola Hierarchical data structures are readily found in survey research. For example, households are found within blocks, cities within area codes and respondents are nested within interviewers. Akhough, here, we are interested in the latter, one can generalize the use of the currently proposed method to similar forms of clustering found in research. Various authors have criticized existing studies on their statistical model failures to account (statistically) for interviewer effects when making inferences based on respondent characteristics (Hagenaars and Heinen, 1982; Dijkstra, 1983; Groves and Fultz, 1985). We show how the SAS macro by Karim and Zeger, updated by Groemping, is utilized in a national behavioral methodological survey/study to control for interviewer effects (clustering of respondents within interviewer) when making inferences based on items influenced by interviewer manipulated study variables. We briefly describe the limitations and give conclusions. regression analysis, macro definition source 2
64 Using the SAS System in the Principles of Economics Course John F. Kileen,
Joseph E. Earley		 The purpose of this paper is to illustrate how the SAS System can play an important part in the teaching of introductory economics at the College and UniVersity level. The authors present a summary of their experience using the SAS System lor this purpose. The paper divides eoonomios into the traditional micro and macro topics and then discusses the various SAS procedures used in these areas. proc step, proc reg, regression analysis source 2
65 Using the SAS System to Produce HTML-Based Text Richard D. Langston The author has previously produced a poster session regarding the presentation of genealogical data. The next step in the process of producing a useful document of research is to use emerging HTML technology in conjunction with the World Wide Web. The author has used the SAS System to produce a set of HTML files consisting of text and image data. The resultant files can be examined by common Web browsers and can also be converted into a serial form suitable for publication. sample, txt source 4
66 Warehouse Now! The Legacy Uncovered Jay A. Jaffe "Data Warehouse" is but a pathway for generating decision-support information from base data sources. If our focus is on getting business information to decisionmakers, then we can construct a system to provide warehousing services to the enduser, even for using Legacy data-- i.e., data stores that were not created as a result of a physical data-warehouse design. By separating the issues of enduser-interface and data management, we can use appropriate technology to generalize the warehouse services and respond flexibly to challenges of client platform and incremental construction. Importantly, we can embark on these efforts at our local department level, without necessarily involving corporate Information Technology. stage 1, stage 2, stage 3, data warehouse, dataset, macro call, sas programmer, raw data, macro definition source 6
67 What's Bugging You?: A Guide to Debugging SAS Programs Susan J. Slaughter,
Lora D. Delwiche		 put statement, missing value, if-then-else, sas log, error message, character variable, input statement, proc step, proc print, dataset source 10
68 Why Knowing About Our Customers Helps Our Business A Practical Application of Business Intelligence Software John McIntyre,
Marc Halsted		 The proliferation of affordable and powerful hardware in modern enterprises, the increase in the number of knowledge workers capable of making valuable use of these tools, and the realization that critical examination of data about the enterprise can really have bottom line impact, have combined to drive modem businesses to examine their decision support infrastructure. Of these business drivers, the most compelling is the connection between business intelligence analysis, and measurable impacts on the activity of an enterprise. This paper will step through a case-study which examines how the use of business intelligence tools from SAS Institute helps identify new opportunities for the enterprise. proc step, proc report, dataset, proc sql, flat file, business intelligence, proc sort, class variable, if-then-else, data science, excel, proc freq source 9

Back to Conf go 1997 go 1995

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.