WUSS 1998 Papers - Index

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

#	Title ×	Authors ×	Abstract ×	Keywords ×	Extras	Pages
1	%FLATFILE, and Make Your Life Easier	M. Michelle Buchecker	%FlA TFI LE is a macro that will create a flat file from a SAS"' data set. This macro takes 3 parameters: the SAS data library name, the SAS data set name. and the name of the flat file to create. The beauty of %FlATFILE is that~ will query the data set to determine the variable names and write out the PUT statement INCLUDING the variables' permanent format (or a defau~ if there is none). Operating Systems: ALL Version: SAS 6.07 or higher	macro variable, macro facility, put statement, dictionary tables, macro loop, character variable, do loop, flat file, proc step	source	3
2	A Competitive Bid Pricing System Using JMP	William R. Gjertsen	Many sales situations require competitive bids before the purchase award is let. From a past history of let awards and knowing only your cost and whether you won it, you can use the inverse prediction option for logistic regression to determine reasonable values to bid.	anova, training set	source	8
3	A Health Care Claims Data Mart: Construction and Exploitation	Stuart B. Levine, Marge Scerbo	Health care claims data are complicated animals. The eccentricities of this data include differences in the types of claims files and the services which populate those files, variations in the type of providers who service the population and the information needed about each type of provider, and finally, the actual populations of health care recipients and all of their .individual specifications. Being able to sort through all of this data (and there is usually a lot of it!) and answers queries from Insurers, providers, analysts, etc., and other people within your organization can be ·a. real challenge. In the past, the solution was to have people submit requests for one or more programmers to extract data and run analyses. The problem was twofold: programmers were overwhelmed with requests, and worse, the data and/or analyses requested were not the correct solution to the problem. This paper will highlight how this challenge was met through the development of a Data Mart.	data mart, dataset	source	7
4	A Non-Linear Calibration Problem in an Enzyme Linked Immunosorbent Assay	Louis Munyakazi	Statistical calibration has generally been concerned with the estimation of an unknown explanatory variable based on an observed response variable. The functional relationship between the variables could be either a simple linear model or it could be through a known non-linear function. The intent of this poster is to present some of the early steps of an immunoassay from the development stage to the validation analysis phase. At the early stage of development, these steps include: (1) an accurate characterization of the response from a calibrated concentration by a non-linear function such as the four parameter logistic curve, (2) an application of statistical procedures to test linear hypotheses about the estimated non-linear parameters, (3) a search for (possible) optimal design that could increase the precision of the estimates, and (4) the derivation of the inverse function for further use in estimation of the analytical concentration in an unknown sample from a calibration curve fit to known standard concentrations.	poisson distribution, regression analysis	source	4
5	A PROC IML Macro for Detecting Health Care Fraud	Jimmy Thomas Efrid	We present a `PROC IML` macro that is based upon the concept of multivariate distances (e.g., d-scores) to identify potentially fraudulent claims in a suspect health care provider organization. The data input is in the form of a p-variate collection of observations from two multivariate populations (e.g., Fraud vs. No Fraud). In addition to producing d-scores, the macro performs Hotelling's T-Test on the p-variate vector of means and calculates simultaneous confidence intervals for the mean subtest differences between the two populations.	confidence interval, proc iml	source	3
6	A Procedure to Automatically Update Web Pages with the SAS System and a Web Agent	Loan X. Nguyen	This pa~r presents an application that uses the SAs* System, a web agent. and a scheduler to automatically update information on the web.	call symput, missing value	source	4
7	A SAS/IML for Guttman Scale Analysis	Dongyun Yang, Chih-Ping Chou	Guttman scale analysis is an analytical approach to examine the underlying operating characteristics of three or more items to determine if their interrelationships meet several special properties that define a Guttman scale. Each item in a Guttman Scale must be dichotomized (e.g., yes/no), and all of the constituent items can be ordered (e.g., from most frequent to least frequent). Guttman scales must be unidimensional and cumulative. This technique has been widely utilized in social sciences research, especially in the area of attitude measurement in survey research. The Guttman procedure has been eliminated from the SAS program. Unfortunately,· no alternative procedures have been made available to SAS users.	chi-square goodness-of-fit, degrees of freedom	source	6
8	Accessing Data from Your PC Using Version 7 of the SAS System	Roger Sanders	Data access technology is changing rapidly in today's computing environment, especiaRy an the PC platfann. The Version 7 release of the SAS System is the first release that contains a newly designed suite of SASIACCESS 8 sottware products that strive to take advantage of recent technological advances in data access. This paper provides an overvieW af the SAS external data access capabilities that are &Vallable an the PC. Emphasis is placed an the `SAS/ACCESS` products, the lmport/Expart Facility. ODBC drivers that are available, and conversion utilities for existing MS SQL SeiYer and ASJ4008 view descriptors.	proc step, dataset, proc sql, oracle, excel, relational database, proc export, proc import, sql pass-through	source	5
9	An Innovative Approach of Analyzing Longitudinal Data ith Various Patterns of Missing Data	Shiao-Ping Lu	Consider a longitudinal study wherein two or more groups of individuals have continuous outcomes measured at baseline and at pre-specified time points thereafter with some observations possibly missing. For comparing two groups, several possible approaches might be used, which differ in terms of informativeness, simplicity, statistical power and assumptions. One approach is to perform an analysis of variance with repeated measures; however, this requires assumptions of normally distributed data and of compound symmetry correlation structure, which may not hold. It also requires an equal number of observations for each person, which is usually not the case in a typical clinical study. Other model-based techniques such as GEE (General Estimating Equations) allow for missing data, but require certain underlying correlation structures, which may not be met. Another approach, and the focus of this paper, is referred to as the summary statistic method described by Dawson and Lagakos. With this method, each subject's response vector is transformed into a single statistic, such as the slope of the least-squares regression of the response variable or area-under-the-curve-minus-baseline.	missing value, proc means	source	3
10	An Overview of Techniques to Introduce External Data Into the SAS System	Andrew T. Kuligowski	The SAS System has numerous capabilities to store, analyze, report, and present data. However, those features are useless unless that data are stored in or can be accessed by the SAS System. This presentation will provide a brief introduction to many of the different methods that can be used to pass data into the SAS System. Topics will include the menu-driven SAS Import Wizard, the INPUT and INFILE statements, DDE, ODBC, and many others - as many as time and space will allow!	input statement, infile statement, character variable, proc sql, filename statement, proc step, column input, list input, variable, excel, file statement, dataset, named input, sample, length statement	source	20
11	Analysis of the American National Election Survey for 1996 using SAS Multivariate Statistical Techniques with Emphasis on Chi-Square	Kamal Shoukry	The purpose of this paper is to show how the SAS System may be used to analyze and contrast private and public views given to researchers in the 1996 National Election Survey of the University of Michigan. SAS procedures are used in order to find out the extent to which public and private views are related.	proc freq, proc step, fisher exact test, exact test, sample	source	4
12	Analysis of the Multivariate ENSO Index Using SAS/ETS Procedures	Joseph E. Earley	El Nino/Southern oscillation [ENSO] is a disruption in the ocean-atmosphere system in the tropical Pacific having important consequences for weather around the globe.' The purpose of this paper is to illustrate how the SAS System may be used to gain insight into the nature of the Multivariate ENSO Index (MEl) which is used in modeling and forecasting ENSO activity. The paper presents the	time series analysis, proc arima, proc step, correlation, proc calis	source	4
13	Automating a Follow-up Clinical Trial Data Collection System Using HP UNIX SAS/SHARE, SAS/AF Frame Entry, and SAS FSP	Long Ngo, Nick Pászty, Christopher R. Cartwright	This paper demonstrates how a clinical trial follow up study was automated through the integration of several SAS system components under the UNIX operating system environment. The system is a server based HP UNIX SAS software based application that incorporates `SAS/SHARE`, `SAS/AF` Frame Entry, and SASIFSP as well as UNIX operating system commands embedded in Screen Control Language (SCL) programs. It is made up of two main parts. Part I, The Data Entry System (DES) is accessed by data entry personnel and stores confidential demographic information received from participating sites as well as additional information retrieved from in-house SAS data sets. These received data are double entered, compared and finalized and for each site, a comma-delimited file containing address information is generated and e-mailed to the investigator. This information is used to produce formal letters, which are sent to the patients informing them of the study and the upcoming interview.	proc step, delimited file, proc print, clinical trial, sas output	source	6
14	Building An Online Clinical Evaluation System Using SAS/AF Frame Entry, and SAS/FS{	Long Ngo, Adam Zhang	Technologies such as 12-Lead Electrocardiography (12-L ECG) and Holter have been used extensively for patient diagnosis in clinical trials and epidemiological studies to help physicians evaluate patient's heart conditions. QT-measurement analysis is one of the important components of 12-L ECG analysis technology, and prolongation of QT interval is known to correlate with life-threatening ventricular arrhythmias. The coding procedure for QT-measurement analysis involves three independent medical coders. The first two coders are authorized to access the online comparison report besides being able to code the patient's QT data, and to view his or her own coded data. Coder three, the consensus coder, is allowed to evaluate, submit, or delete a patient from the consensus database, and to view content of the audit trail database. The SAS software version 6.12 for liP UNIX was used to automate the entire coding system, allowing online coding through Exceed X windows software.	audit trail, clinical trial, sas output	source	6
15	CATALOGER: An Application Development Tool to Search, Compare, and Document SAS Catalogs and Data Files	Christopher A. Roper, Michael Gilman	This paper will present an application written entirely in SAS that greatly enhances the developer's ability to develop, compare, and manage SAS catalog entries and datasets. This application has point and click capabilities for printing catalog entries, including images of frame entries and complete descriptions of associated widgets directly into Microsoft Word documents, creating publication quality documentation at the push of a button. The batch printing facility allows for printing of all catalog entries of virtually any type, including Program, Source, SCL, and Slist (including expanded sublists) entries. The text searching facility will search through all text-based entries for character strings and allow point and click access to the locations in any catalog entry containing the text string.	catalog entry, catalog, dataset	source	6
16	CNTLIN Option: More Than Just Another Pretty Format	Robert O. Bernstein	This paper describes how SAS software was used to match merge data even though the identifiers contained millions of errors. More than 16 million insurance company vehicle records were matched with records from a data set of more than 30 million records from the California Department of Motor Vehicles (DMV). `PROC FORMAT`, with the CNTLIN option, was used to prepare the data for secondary match routines. Vehicle Identification Numbers (VINs) were the primary matches. With the help of the CNTLIN option, VINs that were close (off by one of the 17 characters or those containing transposed characters) were matched. The results included uninsured vehicles by ZIP Codes, counties, type of vehicle and vehicle age.	data step, proc format, proc step	source	6
17	Constants for Two-Sided Normal Tolerance Intervals	William N. Anderson	Two sided normal tolerance intervals are fundamental for the analysis of product reliability. These intervals are of the form x ± ks. For sample size n the constant k is chosen to give the desired confidence that the in terval contains a specified proportion of the true popu lation. The constant k cannot be expressed in terms of standard statistical functions. This paper presents a SAS macro that will compute values for the constants k for any values of the parameters. The computation uses an exact algorithm; in contrast, most published tables use approximations. SAS `PROC CAPABILITY` also ~ses an approximation.	sample, tolerance interval, sample size, proc step, degrees of freedom, data step, macro variable, confidence interval, excel	source	6
18	Constructing a Data Warehouse	Curtis A. Smith	Lately, we have heard and read much about' data warehousing. While many papers discuss the concepts and reasons for data warehousing -here the author will describe methods to build a data warehouse. A data warehouse must be well organized, easy to understand, easy to use, and easy to reach. It will likely draw data from many sources, perhaps for multiple company segments, for multiple time periods, and varying degrees of detail. Should the developer create muttiple SAS System data libraries, or keep all the SAS data sets in a common library? Should the developer use a naming convention for SAS data libraries and data sets? What about documenting the data warehouse? Are indexes a good thing? can indexes be overdone? Subsets and summarized SAS data sets? All these questions and more will be answered in the pages lo follow as the author describes learns how to construct a data warehouse.	libname statement, dataset, raw data, data warehouse, cpu time, proc means, by group processing, macro variable, proc step	source	5
19	Controlling for Group Differences and Counting Rare Events: Propensity Scores with PROC CATMOD and Poisson Regression with PROC GENMOD	Stefanie J. Silva, Lori Potter	In a retrospective study comparing treatments for hypercholesterolemia and their impact on economic and clinical outcomes, the application of two SAS procedures is demonstrated. In this atypical analysis, `PROC CATMOD` was used to create propensity scores to control for systematic differences between treatment groups. We compare this method, which uses patient characteristics to predict group membership, to the more common approach of including a large number of separate covariates in the regression model Analysis of rare clinical events was done using `PROC GENMOD` to perform Poisson regression. We show how we derived adjusted means with this procedure.	proc step, regression analysis, poisson regression, propensity score, poisson distribution, proc genmod	source	5
20	Creating Inferred Data Using SAS	Tim Walters	Because of the costs and difficulty of data collection. companies often do not collect all the data an analyst needs. Base SAS provides a number of tools to create inferences, or "inferred data", from data that is commonly collected. This paper describes various data that can be inferred, procedures for standardizing data, three programming techniques to generate inferred data, and limitations on creating and using inferred data.	index, proc means, sas base, character variable, proc step, proc freq	source	6
21	Creating Instructional Modules Using SAS/IntrNet Software: The Contigency Table Home Page	Stuart E. Wugalter, Jeannie Lum, Colin Campbell	This paper will describe an example of a SASIIntrNet application that dynamically produces two-way tables and their associated statistics upon request from a web page. The SAS application is simply a `PROC FREQ` which returns its results to a web page via the ~oout2htm macro that is a component of the SAS Web Publishing Tools. There are many similar applications on the web, but they · are not necessarily instructional modules. This presentation will show the components and considerations necessary to incorporate such an application into an instructional module.	proc freq, proc step	source	6
22	Data Libraries on Tape	Curtis A. Smith	Big data forces the SAS System developer and usertoturnto big storage media. Often this means - tape. While tape solves the storage space dilemma, it comes at a price. There are limitations, problems, and inefficiencies inherent in tape storage. Within the pages to follow, the author will explore many such difficulties with tape and the ways to overcome the difficulties. For example, tape devices are sequential storage unns rather than random access devices, as are disks. Wihat problems does sequential storage bring? Can multiple SAS data sets be stored within a single SAS data library on tape without destroying each other? What effective SAS capabilities are sacrificed when using tape? How does the user allocate a SAS data library on tape from within SAS? All these questions and more will be answered in the pages to follow as the reader learns how to use SAS data libraries on tape.	dataset, libname statement, proc step, proc copy, error message	source	4
23	Data Tables: Highlighting Techniques	Christopher A. Roper	Data tables are arguably one of the best looking widgets in SAS. This makes them a popular choice for replacing the venerable Extended Table widget as a selection list, or front end to a data set. The modeVviewer paradigm is an extremely flexible tool that gives the Data Table, and it's parent class the Table Editor, an extensive utility across a broad range of applications. Combined with the abundance of methods available to the Data Table class, this widget can add a very nice touch of polish to an application, giving it the professional, finished appearance a client demands and deserves. This paper will illustrate five scenarios for using highlighting to make the Data Table object more attractive and functional for your information consumers.	dataset, if-then-else	source	8
24	Data Warehouses and Bayesian Analysis - A Match Made by SAS	Ram Gopalan	Data warehouses have the advantage of storing and retrieving large amounts of data in a coherent and efficient manner. Bayesian analysis uses large amounts of data either from outside sources or previous analyses. SAS is probably the only software environment that allows seamlessly connecting data management through warehouses and Bayesian analysis. In this paper, I will demonstrate how data warehouses can be used as a tool in tracking prior data and posterior estimates in a Bayesian data analysis model for linear regression. `SAS/AF` will be the front-end tool with `PROC IML` performing the Bayesian Regression model. Key words: prior, posterior, linear regression	bayesian statistics, posterior probability, data warehouse, likelihood function, predictor variable, prior probability, regression analysis, response variable, data mining, proc reg, beta distribution, data science, binomial distribution	source	6
25	Deciphering the _TYPE_ Variable in MEANS and SUMMARY Output Data Sets	Arthur L. Carpenter	Output data sets generated using the procedures MEANS and SUMMARY automatically contain the variable TYPE . This variable can prove to be very valuable ~hen "Creating or working with subsets of the summary data set. but knowing what values of TYPE to select for is difficult if you do not know how itS values are assigned. Fortunately the values assigned to `_TYPE_` depend on the presence of the CLASS statement and the number and order of the variables that it contains. Understanding the relationship between the CLASS statement and TYPE will allow you to accurately predict the va..;8 of `_TYPE_` that will be aSSO!:iated with a particular combination of variables in the CLASS statement. This value can then be uSed to select data subsets based on this relationship. As it turns out this relationship is fairly simple to understand and this paper will show you how to determine the resulting value of `_TYPE_` based on an\t given combination of CLASS variables.	class variable, proc step	source	3
26	Deriving Counts from Transaction Data: Hints for Getting What You Want and Understanding What You Get	Mark F. Meierding	The task of deriving counls from transaction data often appears straightforward, then rapidly becomes complicated. Using insurance policies as an eJCample, this paper presents guidelines for analyzing counting processes and applying appropriate SAS System techniques. Amcng the guideline 1Dpics are: Using `PROC SUMMARY` with multiple BY or CLASS variables. Checking your assumptions about consistent values for ID variables. Merging counls with files that have a longer summary key. Using `PROC SORT` and the RETAIN statement to delermine a count from clusters of related records. When to use negative counls. The time element: counting "s1art" vs. ·as of" vs. "duration·.	proc means, proc step, if-then-else, sample, natural key	source	5
27	Dictionary Tables	Michelle Buchecker	Data about data about data ... The Dictionary Tables provides a weatth of information about your SAS Data Sets. lnfonnation such as the names of the variables, their types, lengths, fonnats and labels. as well as what data sets exist in all libraries. Ever want to know which of your data sets contains a variable named YEAR or which variables have a MMDDYYS. format? Query the Dictionary Tables! Especially handy when application wide changes need to be made. Does the Year 2000 ring a bell? Operating Systems: ALL Version: SAS 6.07 or higher	dictionary tables, proc sql	source	2
28	Dynamically Creating a Where Statement	Curtis A. Smith	Frequently, within an automated application, the developer uses a WHERE statement to subset a data set. The constant values for the WHERE statement can easily be passed from the user using a variety of methods. For example, the developer can use a macro variable that is passed when the user makes a selection from an `SAS/AF` FRAME. However, what if the developer wants to allow the user to make any number of selections? For example, if a WHERE statement is used to select from the account variable, what if one time the user wants to select one account, but the next time the user wants to select ten accounts? How does the developer program a WHERE statement that can accommodate a varying number of constants? Well, thanks to some fancy help from the SAS Institute's technical staff, the authorwill present a handy macro loop to dynamically build a WHERE statement.	where statement, dataset	source	3
29	Estimating Standard Errors of Treatment Effects for Probit Models and for Linear Models of Log-Transformed Variables with PROC IML	David J. Pasta, Miriam Cisternas, Cynthia L. Williamson	Analysts who work frequently with linear models using the SAS System often use the LSMEANS statement in `PROC GLM` or `PROC MIXED` to obtain "adjusted means." The absence of a clear analogue to these least-squares means for nonlinear or generalized linear models can be an obstacle for both the analyst and the eventual audience of the analysis. This paper shows how to use `PROC IML` to estimate standard errors of treatment effects from a probit model and from a linear model on log-transformed data. The probit analysis modeled the probability of intensive care hospitalization; the effect of interest was the difference in average probabilities for two treatments. The linear model on the logarithm of health care costs also compared two treatments; the effect of interest was the difference in cost, not the difference in the logarithm of cost.	proc step, proc calis, regression analysis, proc iml, proc glm, treatment effect, anova, response variable, covariance, proc reg, proc mixed, predictor variable, proc genmod	source	6
30	From Madness to Method	Diana D. Suhr, Thomas J. Gavin, Susan B. Kovitz, Brent S. Goodman	The poster presentation will illustrate a method to display data in a concise and simple format. SAS programming .code using `PROC TABULATE` will be presented as well as the final product, a tabulate table. Actual data compiled by the Office of Institutional Research and Planning at the University of Northern Colorado will be used as examples to illustrate the process of funneling data into an ordered, organized table. SASCode The general SAS code for `PROC TABULATE` (SAS Guide, 1990) is: `PROC TABULATE` DATA=dataset MISSING	proc tabulate, proc step, class variable	source	6
31	Fundementals of FRAME: A Tutorial for Building Applications	Gregory S. Barnes Nelson	SAS!AF 1 FRAME technology allows us to develop user interfaces that are friendly to use. They are the basis for many SAS applications in which end-users determine what happens with the click-of-a-button. This paper presents a framework for building intuitive user-based applications. Here we will develop a simple application, which builds on itself iteratively. By exploring some of the most common· FRAME objects and understanding how events are handled in SAS/ AF, we will build a simple interface for an applications. Through this step-by-step process, we will show you how to build an interface that collects user input, which can be used to front-end traditional SAS macros. Despite it's simplicity, this paper should give the reader a methodology for developing user based applications that can serve to front-end most run-time, interactive macros. This paper is intended for both new users of SAS Software as well as experienced SAS programmers with little or no experience in SAS/ AF.	dataset, sas programmer	source	10
32	Generating Pipeline Chart by Using SAS/GRAPH	Jeff Y. Yuan	Pipeline Chart is a graphical representation of the capability and the centering of parameters that are critical to a process or a production line. This paper illustrates the use of SASIGRAPH to generate the informative Pipeline Chart. In the chart, the horizontal axis has the name of the parameter and the vertical axis is the normalized measurement of all parameters. Two horizontal and parallel lines represent the specification limits of the parameters and another horizontal line, in the middle of the specification lines, represents the target value. For every selected parameter, statistics such as mean, standard deviation, Cp, Cpk, and the failure rate of all data collected during the reported period are calculated and displayed in the chart.	proc step, normalization, proc sql	source	4
33	Getting from Complex Databases to a SAS Analytic File: Tales from the Trenches	Marcia Stoddard	Increasingly, data destined for statistical analysis in SAS come from one or more complex databases. In this paper the author demonstrates various methods of gathering information from multiple, complex data streams and knitting them together into one or more analytic data sets. The types of data considered include date-specific event histories, episodic clinical data, and patient provided panel data. The focus is on how to use the linkages among these various types of information to create analytic files in SAS. This paper also explores how to apply summarization techniques to both prospective and retrospective data sources.	clinical data, panel data, proc freq, if-then-else	source	6
34	Getting Started with PROC LOGISTIC	Andrew H. Karp	Logistic regression is an increasingly popular analytic tool. Used to predict the probability that the 'event of interest' will occur as a linear function of one (or more) continuous and/or dichotomous independent variables, this technique is implemented in the SAS System in `PROC LOGISTIC`. This paper gives an overview of how some common forms of logistic regression models can be implemented using `PROC LOGISTIC` as well as important changes and enhancements to the procedure in Releases 6.07 and above of the SAS System.	logistic distribution, proc logistic, logistic regression, predictor variable, proc step, regression analysis, response variable, odds ratio, estimate, proc reg, confidence interval, null hypothesis, data science, proc genmod	source	5
35	How I Converted a Batch Application System to Client-Server and Lived to Write About It	Curtis A. Smith	After years of using batch applications, even though they were user friendly, my users still wanted something GUI, something that runs on their desktop, and something that interchanges with their other Windows applications. So, I set out to convert my batch application system to something my users wanted. Armed with `SAS/CONNECT`, `SAS/AF`, a TCP/IP connection, Windows 95, my large scale server, my existing batch code, and the will to succeed, I endeavored to build a user friendly, point-and-click, drag and-drop, client-server environment. Within this paper I will describe the steps I took, the tools I used, and the lessons I learned. However, I will not discuss how to write SAS code, SCL, or how to create a SASIAF FRAME; just describe how they are used within my client-server environment.	macro variable, data warehouse, excel, sas base, dataset, txt, data science, proc print, regression analysis, macro definition	source	5
36	How to Get Last Month's DATA "Automatically"	Wayne B. Finley	Frequently, I need to extract last month's data from an operational system and store it in a SAS data set. I then can easily write SAS reports for that month. This paper describe an easy and automatic way to extract the prior month's data from an operational system and store in a SAS data system without having to enter any parameters such as 1st day of the month or the last day of the month.	macro variable, dataset, proc step	source	2
37	Illustration of Improving Kaplan-Meier Plots in the Presence of Heavy Left Truncation	Alison J. Canchola	When performing a survival analysis with staggered entries, subjects enter the risk set at some point after the time origin. Because the Kaplan-Meier survivor function is cumulative, if there Is an early event when few subjects are in the risk set, the early instab~ity can cany forward and cause the entire survival curve to be underestimated. Comparison of two curves can be especially problematic when one curve is more heavily left truncated than the other, as in the case of splitting the data based on a time dependent covariate. We illustrate two possible solutions {as proposed by Bacchetti and Segal, 1995) using SAS" software: (1} left truncating both survival curves at the 20th percentile of the truncation times in the more heavily truncated curve and {2) using a smoothed hazard plot instead.	survival analysis, kaplan-meier plot, hazard function, proc phreg	source	6
38	Improving Performance and Ensuring Scalability of Large SAS Applications and Database Extracts	Mike Beckerle	Many organizations that have invested heavily in data warehouses, data mining systems, and other large business intelligence systems are now discovering that the sheer volume of data being processed prevents large jobs from completing within the batch window. In some notable cases, long runtimes have been crippling--preventing the data warehouse from achieving its goals. In many cases, though, the long runtimes represent an inefficient use of the data warehouse that exacts other significant costs, for example, preventing the organization from running the job more frequently. Many organizations are therefore looking to cut batch runtimes by using multiprocessor systems with parallel processing capabilities-for example, SMPs, MPPs, and clusters. Parallel databases like DB2 UDB makes excellent use of these machines because they partition data across multiple nodes and process all database functions--including insert, update, and delete--in parallel.	dataset, parallel processing, flat file, proc step	source	5
39	Influence of 1997-8 Asian Crisis on Enrollment at a Private Los Angeles, CA University	John L. Rooney	American Intercontinental University is a proprietary school with four campuses. over 60% of the enrollments at the Los Angeles campus are international students. SAS was used in 1996 to project enrollments for 1997 and the forecasts proved accurate. A similar	anova, proc step, regression analysis, response variable, proc reg	source	3
40	Introducing the SAS System for Windows into a UNIX Statistical Computing Environment: Challenges from a User's Perspective	Colin Hessel	Although the SAS System for UNIX is a popular platform for statistical computing, many users expected to work on this platform may prefer to use the SAS System for Windows. This paper examines the challenges associated with sharing SAS data sets, programs, and fonnat libraries when working in a mixed environment Maintaining a single ''reference" set of data and programs may also be difficult Suggested strategies for addressing these challenges include the use of transport files, `SAS/CONNECT`, operating system specific macros and scripts.	proc step, automatic macro variable	source	3
41	Investigating the Performance of Model Selection Information Criteria Using SAS/IML and SAS/MACRO	Long Ngo	Since its discovery in March 1971, Akaike Information Criterion (AIC) has been used frequently especially in time series, and more recently in statistical models for repeated-measures data. However, its properties such as consistency and optimality and underlying assumptions are not clearly understood. To use AIC in model selection procedure, one has to assume that the true model is contained within the families of approximating models; otherwise one needs to examine other AIC-related criteria such as the Kullback-Liebler criterion which is much more computationally intensive and is not implemented in any of the statistical procedures in SAS software. How good is AIC's performance if the above assumption is not valid? If Kullback-Liebler criterion is to be used, what are the computational issues? In this paper, simulation studies will be done using SASIIML and `SAS/MACRO` to try to understand some of these performance issues.	sample, feature selection, sample size, likelihood function, maximum likelihood estimation	source	6
42	It's Almost the Year 2000 - Should I Panic?	Mike Kalt	This paper addresses the most fr!Mluently asked questions about Year 2000 compliance for the SASe System. Topics include: Is the SAS System Year 2000 compliant? What should I look for when testing my SAS applications for Year 2000 compliance? What resources are available to test my SAS applications for compliance? Do I need special licensing in order to do Year 2000 testing? * How do I correct problems in my data and SAS applications? We assume that you have a basic knowledge of the SAS system. and are familiar with general aspects of the "Year 2000 problem".		source	8
43	Macro Quoting Functions, Other Special Character Masking Tools, and How To Use Them	Arthur L. Carpenter	Quoting functions allow the user to pass macro arguments while selectively removing the special meaning from characters such as &, %, ;, ', and ". Most of these functions are not commonly used and are even less commonly understood. Although they are powerful and can even be necessary, usually programming solutions are available that do not require the use of the quoting functions. When are quoting functions needed? When needed how are they used? Which one should be selected? This paper will discuss when to use and when to avoid quoting functions. In addition the discussion will include solutions that avoid the use of quoting functions. Fortunately there are several ways to mask special characters within the macro language. These include quoting functions, o/oDO blocks, restructuring of statements, and the character mask(%).	macro variable, macro quoting function, macro function, proc print, macro put, macro facility, macro loop, %if-%then-%else, proc step, macro definition	source	5
44	Moving to Version 7 in SAS/AF: An Application Developer's Perspective	Steven A. Wilson	Development of SAS/ AF applications in Version 7 will be much different than in the current Version 6 of the SAS System. Even the name of the AF coding language has changed, from Screen Control Language to SAS Component Language, representing a more robust realization of an object oriented development environment. This change will impact all SAS/ AF development groups as Version 7 is installed at sites throughout the globe. MAJARO InfoSysterns, as a developer of SAS-based clinical software that is installed at many pharmaceutical and biotech companies, is acutely affected by any changes in the SAS application development environment.	catalog entry, dataset, error message, sashelp	source	10
45	Multidemsional Data Visualization Tools	Sanjay Mantage, Jim Beamon, Cynthia L. Huffman	Huge amounts of data are piling up in Data Warehouses all over the corporate world. There are valuable trends about customers, clients, users, patients, products, etc. that are buried in the sheer volume of seemingly overwhelming data. Help is on the way! New interactive Multidimensional Data Visualization tools from SAS Institute are designed to help you extract trends and associations in your data, that are beyond the capabilities of traditional2D tools. This presentation will outline some of the features of the Interactive Multidimensional Visualization capabilities that will be included with the Version 7 `SAS/GRAPH` software.		source	6
46	Multiple Regression and Model Selection on Developing a Computer Achievement Model	Leping Liu, George C. J. Fernandez	This paper will present a study in which a computer achievement model was developed by using multiple regression and model selection with SAS system. Four computer attitude variables - enjoyment, motivation, importance, and freedom from anxiety - were examined to be the important predictors of computer achievement. An intermediate variable - time spent on computer - was determined between computer attitudes and computer achievement.	regression analysis, response variable, feature selection, predictor variable, multiple regression, data science	source	6
47	National Comprehensive Cancer Network (NCCN): Integrating SAS Edit Checks into a Web-Based On-Line Data Entry System	Deborah A. Testa, Rebecca Ottesen, Wendy Paul, James Chen, Eva Lepisto, Joyce C. Niland	This paper describes the integration of web-enabling features of the SAS System into an existing clinical information system developed for the National Comprehensive Cancer Network (NCCN). The prevalence of thin-client web browsers has fully enabled remote on-site data entry of clinical research data, even for organizations that need to operate on a not-for-profit budget This project, which requD:ed minor effort but produced significant impact, should encourage anyone who is considering moving a traditional data entry system into the world of browsers and the internet The application will be demonstrated on-line during the presentation.	data validation rules, proc step	source	2
48	New Directions in Documentation, Help, and Computer-Based Training	Brenda Kalt, Gary Meek	With Version 7, the complete library of SAS reference documentation will be delivered online in HTML. In addition, the SAS help system and computer-based training will migrate to HTML as well. This paper accompanies a demonstration of SAS OnlineDoc TM for Version 7. Examples from SAS OnlineTutorTMfor Version 7 and the SAS help system in HTML will also be shown. The demonstration will highlight new features of the interface for these products and give tips for getting the most out of them at your site. Solutions@Work'"", a continuing series of CO-ROMs that focus on task-oriented usage information with downloadable examples, will be discussed briefly.	catalog entry, date variable	source	4
49	New SAS Procedures for Analysis of Sample Survey Data	Anthony B. An, Donna L. Watts	Researchers use sample surveys to obtain informa tion on a wide variety of issues. Many surveys are based on probability-based complex sample designs, including stratified selection, clustering, and unequal weighting. To make statistically valid inferences from the sample to the study population, researchers must analyze the data taking into account the sample de sign. In Version 7 of the SAs System, These new procedures are being added for the analysis of data from complex sample surveys. These procedures use input describing the sample design to produce the ap propriate statistical analyses. The `SURVEYSELECT` procedure selects probabil ity samples using various sample designs, includ ing stratified sampling and sampling with probability proportional to size. The SURVEYMEANS proce dure computes descriptive statistics for sample sur vey data, including means, totals, and their standard errors.	sample, proc step, regression analysis, proc surveymeans, sample size, random sampling, sample weight, proc surveyreg, standard deviation, proc surveyselect, dataset, confidence interval, proc calis, covariance, degrees of freedom, proc genmod, proc means, predictor variable, output delivery system, cluster sampling, proc glm	source	11
50	Overview of Java Components and Applets in SAS/IntrNet Software	Barbara Walters, Donald R. Chapman	This paper describes the· Java coml?,llnents and the sample applets that are part of SASIInt!Net . software. This paper describes how to use the SASISHARE"' driver for JOBC SASICONNECT"' driver for Java, and the Tunnel feature to access inf~ation and sefVices available from SAS" servers, and also prov1des code examples to demonstrate these capabilities. Note that the code examples referenced in the text are presented in a separate section at the end of the paper.	proc sql, bootstrap, proc step, proc download, select	source	9
51	Preparing SAS Software Applications for the Year 2000	Kirk Paul Lafler, Sunil Kumar Gupta, Charles E. Shipp	Solving the Year 2000 century date problem will be the largest computing task ever performed in history... and the costliest. Recent estimates for fixing the problem range between $300 billion and $600 billion and are rapidly climbing. Due to the enormous amount of software code that needs to be converted and the limited amount of time before the next century, organizations around the world are rushing to become Year 2000 compliant. Programmers code programs differently, so there is no automated solution or 'silver bullet' to fix the problem. The Year 2000 is not a technological problem, but a business problem with strategic implications that have direct impact on an organization's profitability. Solving it will require a common sense approach that keeps applications working while controlling costs. This paper and presentation will describe the problem, issues and challenges, the SAS environment, what a SAS date is, and SAS System solutions to the Year 2000 problem.		source	5
52	Querying the Data Warehouse with SQL SELECT Statement	Kirk Paul Lafler, Sunil Kumar Gupta, Charles E. Shipp	To appreciate the wealth of information a data warehouse has to offer, you need to understand how to communicate with it. Information is typically accessed using an SQL query. The statement responsible for querying the many tables and returning the requested results in a data warehouse enviromnent is the SELECT statement. It is an indispensable piece of the data retrieval process. This paper looks at the structure of the typical data warehouse, the types of information stored in it. and how users access and retrieve data from it.	proc sql, data warehouse, select, proc step, group by, order by, proc print, index	source	5
53	Quick Results from Statistical Analysis - An Alternative Approach to the Popular Menu Interface Systems	George C. J. Fernandez	A comprehensive application oriented graphical data analysis approach to perform introductory statistical	data science, macro definition, hypothesis testing, sample, dataset, excel, raw data, sas output, chi-square test, probability distribution, standard deviation, statistical software, proc freq	source	8
54	QUIZZER: A Fun, Fast Paced Game	Christopher A. Roper	Quizzer is a fun, fast paced game in the format of the popular TV game show Jeopardy!. Written entirely using `SAS/AF` with Frames and SCL, this application offers the opportunity to have a little fun while demonstrating the range of capabilities in the SAS system. The questions and answers are all related to the SAS system and are geared to be challenging enough for even the most experienced SAS expert, and yet simple enough for novice SAS users to be able to participate. There are two rounds of twenty-five questions, and a fmal quizzer round of one question where a team can bet it all. Contestants test their knowledge of SAS in a competitive, yet friendly environment of fair play, and unwittingly may learn a little more about the SAS system in the process. Why A Game? Quizzer was developed to help organizers of SAS related events in generating interest for the event. An event might be a local users group meeting, a training session, or any place SAS professionals might gather.		source	2
55	SAS and the World Wide Web: Simplified	Faith Reneé Sloan	Corporate use of the Internet has grown rapidly in the past few years. With that growth has come a demand for applications and databases that both internal and external intranet users can access easily and securely. It is now quite possible to develop a functional Web site, complete with database applications, on a desktop computer using the SAS System in a matter of hours or a few days - and often with only a small financial investment.	proc tabulate, ods html, dataset	source	8
56	SAS Institute Quality Partners - Who Are They?	Loretta Jo Schlatzer	The Quality Partner Program was established in the Spring of 1995 and announced at SUGI '21 in Orlando, Florida. The goal of the program at that time and through today remains the same: to support and assist independent consultants whose primary business is providing SAS consulting services and SAS expertise to the user community. This program has grown to include over 180 consulting organizations in the United States and over 30 in Canada. A similar program in Europe brings the total number of Quality Partners worldwide to approximately 250.	quality of service, customer satisfaction	source	6
57	SAS Macros Beyond the Basics	Jay A. Jaffe	So you got some `%LET` statements, some &symbols, and maybe some %MACROs that help you do your work? Let's go ahead and leverage the full power of the SAS Macro facility. After you complete this Tutorial, you will understand the mysteries of &&double and &&&triple ampersands, macro functions, the SYMPUT call you've always wanted to use, Macro Language advances in SAS 6.11/6.09E and Version 7, and more.	macro variable, macro facility, macro put, %if-%then-%else, proc step, macro definition, macro trigger, macro loop, macro function, macro quoting function	source	10
58	Shattered Windows! Possibilities and Pitfalls in Transferring Data Among Fragmented Operating Systems and Different Versions of SAS, EXCEL, and ACCESS	Cecille R. Birner	Imagine sharing data from sources ranging from monstrous mainframes to Filemaker pro on Macintoshes. This paper covers tips to speed transfers, pitfalls to avoid, protocols to set, and illustrates simple ways to transfer data sets when you do not have the usual tools such as `PROC ACCESS` or software to convert data for you. Can you get your data though Windows? Is the window half open or half shut? Come to this talk and decide for yourselfl Outline: I) Introduction Ia) Scope of this paper I b) What is data sharing? 2) MTC's data Sharing Needs 3) General rules for data sharing 3a) Receiving a data set 3b) Naming the file 3c) Variable names 3d) Variable types 4) Four methods for data transfer 4a) Conversion software 4b)ODBC 4c) SAS options 4d) ASCII	dataset, excel	source	6
59	Software Benchmarking under NT 4.0: A Comparison to IBM/MVS Mainframe	Kunling Lu, Beimin Ni	Relatively new in running SAS Release 6.12 under Window NT 4.0, Care Management Institute of Kaiser Foundation Health Plan, Inc., launched a benchmark test Three NT workstations were set up with SAS release 6.12 installed on each hard drive. A remote NT server stored the 1.6 GB testing data. There were two tests. One was to read the 1.6 GB into a 33 MB subset SAS file. The other was to test commonly used statistical procedures. Two testing scenarios were established. IBMIMVS mainframe was used as a baseline. Remote 'Client/Server' processing was found superior to 'Client' processing, yet still fell behind IBMIMVS mainframe performance. Bottleneck problem was solved by using array drives. Network connectivity was found a performance factor. Concurrent users were found to degrade performance even though more processors were added onto the server.	proc step, proc means	source	4
60	Some Utility Applications Using the Dictionary Tables in PROC SQL	Jack Hamilton		proc step, proc sql, macro variable, character variable, catalog entry, create table	source	6
61	SQL Query Window Classes: How to Incorporate the SQL Column Experession Builder Class in Custom SAS/AF Applications	Christopher A. Roper	SQL (Structured Query Language) is the standard language used for querying all major databases, including SAP/R3, SYBASE, ORACLE, RED BRICK, DB2, and others. The SAS Query window creates SQL expressions using `SAS/AF` Classes that are available to the `SAS/AF` developer. One of these class entries is the Column Expression Builder Class, QWCLEXPR.CLASS in the SASHELP.SQL catalog. This is the class used by the Query window to build columns in SQL. This class may be called by any `SAS/AF` entry to present the same interface the SAS Query window uses as a column expression builder. This paper will cover what the Column Expression Builder Class is, how to incorporate it into a `SAS/AF` application, and the nuances of using this class effectively.	proc sql, proc step	source	6
62	SQL Query Window Classes: How To Incorporate The SQL Where Expression Builder Class In Custom SAS/AF Applications	Christopher A. Roper	SQL (Stl'Uctlnd Query Language) is the standard language used for querying all nugor databases, including SYBASE, ORACLE, RED BRICK, DB2, and others. The SAS Query window creates SQL expressions using `SAS/AF` Classes that are available to the `SAS/AF` developer. One of these class entries is the Where Clause Builder Class, QWWHERE.CLASS in the SASHEIP.SQL catalog. This is the class used by the Query window to build SQL where clauses. This class may be called by any `SAS/AF` entry to present the same interfac:e the SAS Query window uses to build where clauses for SQL queries. This paper will cover what the Where Clause Builder Class is, how to incoiporate it into a SAS/ AF application, and the nuances of using this class effectively.	where statement, proc sql	source	6
63	Successful Data Warehousing for Telecommunications Business Solutions (The Philos Method)	Demetris Papaiacovou	The telecommunications industry is rapidly becoming the largest user of high volume information systems. This is especially true in the United States, where telecomnnmications, the 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 that transcend the boundaries of traditional 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.	data warehouse, data cleaning	source	6
64	Tapping the Power of PROC TABULATE for Analysis Reports Clients Understand	Ed Jensen	The reporting capabilities of `PROC TABULATE` are very powerful. In just a few lines of code, a very detailed analysis can be reported. Knowing how to tap into the power of `PROC TABULATE` begins with an understanding of the fundamental structure of a TABLE command. From that basis, more detailed reports can be presented. This paper follows that approach. It begins with an overview of TABULATE, laying down the basic structure of a TABLE statement and how it relies on the CLASS and V AR statements. It then demonstrates by example how more detail can be added to the report by creative use of TABLE basics and the addition of formats, appearance options, and some tricks to enhance the report presentation. The statistics N, SUM, MEAN, and the sometimes tricky PCIN and PCTSUM are discussed. Once understood, TABULATE can be quickly coded to generate reports that are easily modifiable for future analysis needs.	proc tabulate, class variable, proc step, missing value, group by	source	10
65	The Capabilities of PROC FACTOR to Conduct an Exploratory Factor Analysis	Tyler C. Smith		factor analysis, correlation, proc step, proc factor, factor loading, proc score, univariate analysis, proc princomp, regression analysis, scree plot	source	6
66	The Central Limit Theorem Home Page	Stuart E. Wugalter, Jeannie Lum, Colin Campbell	The Central Limit Theorem (CL T) is arguably one of the most fundamental concepts in statistics. This paper will illustrate an interactive `SAS/INTRNET` application which clearly illustrates the theorem through simulation and presentation of results in both textual and graphical displays. Users are able to input parameter values in a simple HTML form, the `SAS/INTRNET` application runs the simulation and displays the results. Teaching statistics offers many challenges to both students and faculty. Creating demonstrations of the CL T for statistics students is often labor-intensive and difficult to implement for individual faculty members. In addition, students often fmd such applications difficult to use, dry and therefore, may not understand the necessary theorem concepts. The CLT Home Page offers a solution that off-loads the aforementioned labor intensive tasks onto a centralized, easy and fun to use World Wide Web application.	sample, central limit theorem, macro variable	source	5
67	The Challenges of GUI Testing	Dawn Amos, Stuart Austin, JoAnna Longman, Allen Malone, Sue McGrath, Jim McNealy, Natalie Morgan		multiplicity, quality assurance	source	8
68	The Elements of SAS Programming Style	Frank C. DiIorio		proc step, missing value, proc sql, macro variable, out=, put statement, proc genmod, if-then-else	source	5
69	The Problem with NODUPLICATES, Continued	Jack Hamilton		proc step, proc sql, proc sort, out=	source	5
70	The Programmer's Solution to Import/Export Wizard	Lora D. Delwiche, Susan J. Slaughter	Do you like what the Import/Export Wizard does for you, but you are not really a GUI person? If the answer is yes, then there is good news for you. Version 7 of the SAS System has two new procedures: IMPORT and EXPORT. The `IMPORT` procedure can read your `CSV` and tab delimHed files, and if you have `SAS/ACCESS` to PC File Formats licensed, you can also read dBASE, `EXCEL`, and Lotus files. The `IMPORT` procedure also supports Microsoft Access databases. The `EXPORT` procedure does the opposHe of IMPORT, converting your SAS data sets into other file types. No more clicking your way through window after window, now you can import and export data with just one or two lines of good old fashioned SAS code.	proc step, dataset, proc export, proc import, excel, delimited file, proc print, sas log	source	6
71	The SAS System under Windows, Version 7	Jennifer Bradshaw Clegg, Gary Mehler	Version 7 of the SAS System for Win dows provides support for Windows 95 and Windows NT 4.0. With this version, the SAS System continues to improve in the areas of user interface, application integration, and setup. This paper dis cusses both new features and enhance ments in these areas. Additionally, this paper also discusses the support pro vided by the SAS System for Windows	excel, proc tabulate	source	10
72	The Use of External Software to Import Data into the SAS System	Richard O. Smith, Arthur L. Carpenter	A brief introduction is presented in the use of external software, such as DBMS/COPy"' and DBMS/Engines by Conceptual Software, Inc., to import external data into the SAS System. DBMS/Engines can be used in Iibname statements to allow easy access to external data in a	dataset, flat file	source	4
73	UNIX Explorer: Pretty Remote Computing	Mark T. Stetz	UNIX Explorer is a SAS/ AF FRAME Enlr)' application that runs under The SAS System for Windows, Release 6.12 and provides a point-and-click interface for accessing files on a remote UNIX computer. The application not only lets a user navigate the remote directory structure, but also manipulate any file on d:te system. Because UNIX Explorer was designed with and for SAS, d:tere are· also several features specifically inteoded to operate on SAS data sets. UNIX Explorer exploits the capabilities of SASIAF and `SAS/CONNECT` and integrates seamlessly into a PC SAS session to provide a robust, user-friendly, and, some might say, fetching interface for remote computing.	proc step, feature selection, where statement, dataset, proc download, sas output	source	9
74	Ups and Down-sizing: Moving a SAS Application from MVS to UNIX	Tracy A. Cermack, Kimberly J. LeBouton	Downsizing in the current business world is deemed necessary for companies to remain competitive. Companies are downsizing mainframe applications to smaller platforms in the hope of saving money and gaining productivity for both users and developers. But is this goal being achieved? This paper explores the realities of moving an entire user department, and their SAS applications, from MVS to UNIX at a major manufacturing company. It explores the stumbling blocks encountered during the project; reviews and discusses the non-technical and technical aspects of the project, including data downloading issues; and provides an update on the more recent adventures and experiences in the implementation of this project, which was first presented at SUGI 22.	dataset, proc step, proc cimport, proc cport, error message, sas programmer, macro variable, sas base, elapsed time, cpu time	source	9
75	Use CALL SYMPUT for More Informative Titles and Footnotes	Lynn G. Palmer	Descriptive information available when your program runs can be placed into titles, footnotes or output using CALL SYMPUT. This feature can be very helpful when running reports routinely. Instead of changing dates, counts, or file names in your code, the program can change them for you. This paper will discuss the use of CALL SYMPUT for gaining access to generated information and will show examples of using this information.	call symput, data step, proc step	source	4
76	Use of Data Enterprise Miner & Multivariate Statistical Analysis to Build Fraud-Detection Models	Shiao-Ping Lu	The increasing number of checking account fraud cases and the dollar loss reported at Wells Fargo Bank evoked the need to build an early-detection program at the point of transaction. Since the fraud rate is relatively small compared to millions of transactions per month, it is both desirable and warranted to explore new predictive techniques and to provide early warnings to the fraud prevention unit. This paper will address data sampling, variable construction and model construction. Several approaches taken in the modeling with the utilization of Data Enterprise Miner include logistic regression, tree-analysis and neural network. The accuracy of prediction, misclassification costs and robustness of various models will be compared. Other multivariate statistical methods such as factor analysis in variable dimension reduction will also be discussed.	proc hpsplit, data mining, neural network, logistic distribution, type i error, data cleaning, regression analysis	source	7
77	Use of Multivariate Techniques to Solve an Urgen Real-life Problem in Pharmaceutical Manufacturing	Karl Glaser	Batches of raw material were, with increasing frequency, yielding finished product which was unacceptable. Investigation revealed that all of the raw material met all existing specifications, and the processing parameters had not changed. Univariate methods were used to try to predict batches which would give rise to failing finished product, with little success. Multiple Discriminant Analysis was then used to sort the raw material into bad and good batches.	proc step, excel	source	3
78	Using Excel: A Dynamic Approach to Report Writing	Matthew L. Greathouse, Karl Jensen	Since its inception, SAS has been a powerful tool for creating statistical reports. For the most part, however, these reports have lacked visual appeal. New products from SAS have been developed to substantially improve the look of reports. While this has proved most useful, it requires the purchase of additional SAS products. Another option is to use the Dynamic Data Exchange (DDE) link which is native to Base SAS software. DDE allows users to "exchange" data between different windows applications, rather than transcribing output from SAS to another software package in order to create the desired report. Transcribing is not only time consuming but also fosters transcription errors. While the basics of using DDE to `EXCEL` have been documented within the on-line help and in "SAS Companion for the Microsoft Windows Environment," this paper will expand this documentation and show how to customize reports by using DDE calls and simple macro statements.	excel, macro variable, macro definition, filename statement, put statement, macro function, sample	source	5
79	Using Macros to Maintain Program Flexibility and Integrity When Accessing Very Large Datafiles	Ed Jensen	When writing programs to analyze large datafiles great care must be given to program flow and the accuracy of the procedure otherwise, considerable time is lost in reruns. Macros provide an excellent way to control the program flow and make last minute modifications feasible by allowing changes to be accurately reflected in all relevant areas. This paper will present the use of macros from creating the macro environment with appropriate system options and the Autocall Facility through making the macros flexible with use of both local and global macro variables as well as macro program statements. Various techniques for managing variable lists, standardizing procedures, saving selected macros in autocalllibraries, self documenting one time changes, and debugging techniques will be demonstrated. Sample standard macros that have been found useful will be shown.	variable list, macro variable, macro definition, macro loop, %if-%then-%else, macro call, macro parameter, sample, %macro parameters, macro function	source	8
80	Using PROC IML to Rank Scores	Karen Walker		proc iml, proc step, correlation	source	4
81	Using PROC SQL and an Outside Table/File with a Database	Lynn G. Palmer	Often, not all the data needed for analysis is in your database. With the `SQL` procedure, outside files/tables can be linked to the tables in the database. This paper will discuss different methods for linking files to databases and the efficiency of each method. The databases used in this discussion are DB2 in an MVS mainframe environment and MS Access 97 in a Windows 95 environment. However, the strategies presented are applicable to other databases and environments.	proc sql, proc step, macro variable, create table, dataset, macro put	source	5
82	Using PROC VARCLUS to Examine Geographical Patterns in Biological Time Series Data Along the California Coast	Stephen Schroeter, John Dixon, Richard O. Smith	Weekly settlement of red and pwple sea urchins (Strongylocentrotus .franciscanusand S. pwpurntus) was measured from 1990 to 1993 at sites in southern and northern California, a period that encompassed El Nino oceanographic conditions. `PROC GLM` and `PROC` V ARCLUS were to elucidate differences in the time series of settlement as used to among study sites. These analyses indicated geographically distinct of sea urchin settlement to El Nino oceanographic conditions.	proc step, time series analysis, correlation, proc princomp	source	4
83	Using SAS to Implement Jackknife and Bootstrap Methods for Survey Data	Tony Ye, Ruby Lin	Analyzing data collected from surveys usually requires using special statistical packages (such as SUDAAN). Although one can use SAS to generate point estimates, the variance associated with the estimate is usually overestimated because of the complicated survey structure. Development of bootstrap and jackknife theory provides an empirical solution for analyzing survey data. Fast growth of computing power made it feasible to use these techniques. We developed a SAS macro to analyze the National Health Interview Survey (NHIS) and compared our results to the one obtained by using SUDAAN. For simple means, this SAS macro replicates SUDAAN's results. For more sophisticated statistical modeling, this SAS macro enables users to fit a larger range of models than using SUDAAN or using any other survey analysis package. In this article, we will also discuss how to use this macro to implement various statistical models to other national surveys.	sample, proc calis, proc step, bootstrap, macro definition	source	4
84	Using SAS Window to Dynamically Manage Your Routine Programs	Linda Gau	Have you ever faced the problem that you have to run several related SAS programs monthly or quarterly, but sometimes you just need to run only part of the routine programs. One of the most important aspects of the SAS Window is its ability to create a user-friendly customized window for a novice SAS programmer to run routine programs dynamically. You may get results from a computer without necessarily having any particular programming skill.	%if-%then-%else, macro variable, macro definition, pointer control	source	3
85	Using the SAS System and Dynamic Data Exchange to Populate Microsoft Word Document with Text, Tables, and Graphs	Mike Harris	Dynamic data exchange (DDE) is one of the means by wbicb appropriately written Microsoft Windows applications can communicate and share data with one another. DDE is a cliellt/server architecture. The application that controls another via DDE is the client, and the controlled application ('Mllch usually provides some kind of functionality not present in the cliellt) is the server. The SAS `DATA STEP` Language can encapsnlate WoniBasic statements and enable SAS programs running on personal computm to be efl'ective cliellts and control the actions of Microsoft Word for Windows (Word} in ways that are useful for creating, populating, and formatting tables, and placing text sttings and graphical elements at predetennined locations within documents.	sample, macro definition, data step, %if-%then-%else	source	6
86	Which Path with PROC CALIS?	Diana D. Suhr	This presentation gives an overview of statistical techniques used in a structural equation modeling analysis (Hoyle, 1995), illustrates three models, and provides the SAS programming code using `PROC CALIS`. Archival data will be used to illustrate a path analysis model (Hatcher, 1994), a confirmatory factor analysis model (Byrne, 1994), and a latent growth curve model (Muthen, 1997; McArdle and Epstein, 1987). Results from the analysis of each model will be compared and discussed. Models This presentation will discuss three types of models that can be tested using `PROC CALIS`. The first model, a confirmatory factor analysis model, investigated the factor structure of a twenty-seven item math attitudes scale.	estimate, factor analysis, confirmatory factor analysis, covariance, proc calis, sample, proc step, latent variable, correlation, standardized residual	source	7
87	WIN100IN for Year 2000 Window Installations	Charles E. Shipp	The procedure presented, WIN1 OOIN, is JCL and SAS code that can automatically install either Fixed or Sliding 1 00-year Windows. It is particularly useful when applied to (pds) partitioned data sets that contain your SAS program source-code libraries. It can also be modified for TSO or CICS to interactively and automatically install windows in SAS programs stored in sequential data sets. For Fixed Windows, the user specifies the pivot year, for the starting year of the 1 00· year window, and for Sliding Windows, the user specifies the years back to position the window; it moves each year. A waming message is added, to begin appearing when it is time to archive data that will otherwise move out of the 100-year period. WIN1001N reads a partitioned data set that contains SAS programs. one program per member, and creates a new pds with a fixed or sliding 1 00-year window installed in each program.	sysdate9, proc step	source	5
88	Working with SAS Date and Time Functions	Andrew H. Karp	Many SAS applications require that operations be performed on data collected in the time domain. Among these types of operations are: determining the frequency that a phenomenon of interest occurs in time; determining the time interval which has elapsed between two phenomena (e.g., length of stay between admission and discharge from the hospital); conditional operations on observations in a SAS data set based on date and/or time values; aggregation of observations from one time frequency to another (e.g., daily to monthly); and interpolation (i.e., estimation) of lower frequency observations from data collected at a higher frequency (e.g., estimating monthly values from data collected quarterly). The SAS System Software provides a wealth of tools for users who need to work with data collected in the time domain.	sysdate9, date variable, datetime variable, datepart, time variable, raw data, numeric variable, dataset, proc freq, proc step, time series analysis, if-then-else, mdy, datetime, sas function, time, subsetting if	source	7

Back to Conf go 1999 go 1997

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.