WUSS 1997 Papers - Index

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

#	Title ×	Authors ×	Abstract ×	Keywords ×	Extras	Pages
1	A Data Management System for Use in Smaller Biotech Companies	Ellen Krantz	It is nearly a given that pharmaceutical/biotech companies use SAS for their clinical trial data analysis. What each company needs to decide is what software to use for data management. Clinical trial data is fairly complex, both in its structure and content, and drug regulatory agencies often require data to be double entered, and to have an audit trail of changes made after data entry is complete. Many large companies have elected to use off-the-shelf data management systems that are not written in SAS. These systems are fairly expensive and require non-SAS expertise for their setup and maintenance. This paper is directed at smaller companies, for which an argument can be made for using SAS for data management, as well as analysis. The biggest reason is that the company already has SAS software and programming expertise in-house. Also, data cleanup is simplified, since complex edit checks are often written in SAS, regardless of the data entry language.	numeric variable, proc step, character variable, data governance, proc compare, audit trail, missing value, adverse events, macro variable, macro definition	source	5
2	A Fast and Robust Method for Pattern Matching in Time Series Databases	Eamonn Keogh	The problem of finding patterns of interest in time series databases (queI)' by content) is an important one, with applications in virWally every field of science. A variety of approaches bave been snggested. These approaches are robust to noise, offset translation, and amplitude scaling to varying degrees. However, they are all extremely sensitive to scaling in the time axis (longitudinal scaling). We present a method for similarity search that is robust to scaling in the time axis, in addition to noise, offset translation, and amplitude scaling. The method has been tested on medical, financial, space telemetry and artificial data. Furthennore the method is exceptionally fast. with the predicted 2 to 4 orders of magnitude speedup actually observed. The method uses a piecewise linear representation of the original data. We also introduce a new algorithm which both decides the optimal number of linear segments to use, and produces the actual linear representation.	time series analysis, distance measure, goodness of fit, raw data	source	6
3	A General Purpose Macro to Obtain a List of Files	Mel Widawski	Would you like to be able to accomplish any of the following tasks? Access and convert all .dbf files in a given directory. Write a SAS macro to maintain an annotated list of files all files in a directory. Create LlBNAME statements for all transport files in a directory. Process all files of a given type in a directory for any reason. In order to accomplish any of these tasks you would need to obtain a list of files. This macro produces a macro variable w~h a list of all of the files in a directory. It may be further restricted to produce a list of files of a given extension in that directory. In the course of presenting this macro the following techniques will be covered:	macro variable, macro call, macro parameter, variable, %if-%then-%else	source	6
4	A General System for Custom Conversion of dBase Data into SAS	Mel Widawski	The aim of this system is to import the tables from a dBase database system as automatically as possible under adverse conditions. When the entry system is constantly changing and in control of another, you need a convenient way to bring in all available tables. Besides dOing this, the system allows for the substitution of lIariable names, special transformations, and conversion of look-up tables to SAS" formats. It is completely modular, and allows for an environment where multiple programmers may be working on importing the same data. The program runs regardless of which modules are present. Switching from a test environment to the production, and back, requires only a single string change in the driver. A mapping of variables to formats across data sets is also produced.	dataset, format statement, macro variable, macro call, macro quoting function, macro definition, macro function, macro loop, macro conditional	source	6
5	A Microcomputer Program (SF-36.EXE) that Generates SAS Code for Scoring the SF-36 Health Survey	Ron Hayes, Cathy D. Sherbourne, Karen L. Spritzer, Wil J. Dixon	This paper describes a microcomputer, SF36.EXE, that generates SAS code for scoring one of the most widely used measures of health-related quality of life today, the SF -36'1'M Health Survey. The SF -36'1'M Health Survey measures eight health concepts: physical functioning, bodily pain, role limitations due to physical health problems, role limitations due to personal or emotional problems, emotional well·being, social functioning, energy/fatigue, and general health perceptions. The generated SAS code derives the 8 SF -36'1'M scales as well as the SF ·36'1'M physical and mental health composite scores. In addition, the program produces code that provides US general population normative scores, age and gender adjusted to one's sample. The significance of the difference between the sample and the general population on each SF-36scale score is also generated. Example input and output files are included.	sample, standard deviation	source	5
6	A SAS Macro for an Endogenous Feedback Model to Study Contextual Analysis	Chih-Ping Chou, Donyun Yang	Contextual analysis studies the impacts of surrounding social context on individuals. This	response variable, macro definition	source	6
7	Advanced Macro Topics: Utilities and Examples	Arthur L. Carpenter	Numerous utilities have been created using the SAS Macro Language and many of the newer utilities take advantage of recent changes and enhancements to the SAS System. The examples of macro utilities presented in this paper include the use of: SASHELP views; `DATA STEP` and SCL functions using %SYSFUNC; the AUTO CALL library; and system supplied AUTO CALL macros. The macros detailed in this paper include examples of: copying members of a catalog; subsetting data sets; working with lists of data set variables; counting observations; and converting macro text to lower case.	macro variable, macro facility, macro definition, call symput, macro function, %if-%then-%else, proc step	source	6
8	An Introduction to SAS/FSP Software	Terry Fain, Cynthia Gareleck	`SAS/FSP` is a set of procedures used to perfonn full·screen interactive data entry, data editing, data retrieval, and data query. This tutorial provides a basic ovelView of SASIFSP software under version 6.12 of the SAS System. Topics inclUde: creating a new SAS data set customizing data screens data entry data query * browsing SAS data sets The focus of this tutorial is on `PROC FSEDIT` although other procedures such as `PROC FSBROWSE` and `PROC FSVIEW` will be briefly discussed. Screen control language (SCL) is considered an advanced topic and will not be covered in this Morial.	proc step, dataset	source	5
9	Analysis of Designed Experiments Using SAS PROC GENMOD	Douglas C. Montgomery, Connie M. Borror, Sharon L. Lewis	There are many industrial situations in which the response variable is non-normal. Traditionally variance stabilization transformations are made on the non-normal response to obtain the properties needed to use ordinary least squares (OLS) and analysis of variance (AN OVA). However, a significant bias in parameter estimation can result; at times, leading to inaccurate and misleading analyses. Additionally, there is the issue of inaccuracies in the inverse transformation made to return the parameter estimates to their original units. Generalized linear models implemented through SAS `PROC GENMOD` offer an alternative to the use of data transformation. This paper demonstrates the use of generalized linear models with `PROC GENMOD` and compares models built with `PROC GENMOD` to those built with data transformations. The measure of performance used to compare the models is the resulting confidence interval about the mean stated in original units.	proc genmod, proc step, goodness of fit, poisson distribution, confidence interval, proc reg, likelihood function, data transformation, anova, response variable, residual analysis, feature selection	source	6
10	Analysis of Student Minimum Wage Perception Using the SAS System	Maria Lizet Arellano, Pedro Mercado, John F. Killeen	The purpose of this paper is to illustrate how the SAS System may be used to analyze questionnaire data. A questionnaire was developed to detennine how college students view the level and desirability of the current minimum wage. Approximately 400 questionnaires were distributed to college students at Loyola Marymount University, a private Catholic institution, and state funded universities and colleges. The Chi-Square `PROC FREQ` is used to test the impact of gender, ethnicity, political affiliations, and income on student perceptious. Format, label, and if then assignment statements are used to consolidate information. Several uses ofPROC REG and `PROC PLOT` are presented in the analysis.	regression analysis, predictor variable, proc freq, proc freq chisq, response variable, proc step	source	4
11	Analysis of Time Series Data - An Application in Health Care	Sujaya Parthasarathy	A time series refers to a sequence of observations that are ordered by time. In time series analysis, one of the important assumptions of classical linear regression - independence between the error terms - is almost always violated. This causes the standard errors of the parameter estimates to be biased, which in turn renders the t-test invalid. Therefore estimation of time series data requires a different approach from traditional regression analysis. This paper discusses some of the important concepts that are peculiar to time series data, and the tools for identifying and estimating the process underlying time series data. Data on the daily perceptual speed scores of a schizophrenic patient are used to demonstrate the use of some of these methods. The SAS system's `PROC ARIMA` was used in this analysis.	time series analysis, moving average, time series data, estimate, residual, stationarity, sample, confidence interval, partial autocorrelation, proc calis, regression analysis, proc step	source	6
12	Basics of Using SAS with the World Wide Web	Marvyn M. Lindsey, Charles E. Shipp	This tutorial will provide an overview of the various techniques that may be employed to access SAS data sources from the W orId Wide Web. The authors will discuss the available tools to access and present SAS data from the World Wide Web including the SAS ODBC driver and the SASJIntrNet components. The advantages and disadvantages of each	sas output, ods html	source	10
13	Becoming a SAS Master	Ginger Carey, Helen D. Carey	The scope of SAS products and their continual enhancement prevents anyone person from mastering the entire SAS System. However, we can become masters in the portions of the SAS System that we use. It is not enough to know the building blocks of SAS - the statements, syntax, and concepts. We also need to know how to put them together to accomplish our work. In this paper, we guide you to resources available and skills needed in order that you can learn faster and better on your way to mastering SAS and the Information Age.	sample, data warehouse	source	8
14	Better Titles: Using The #BYVAR and #BYVAL Title Options	Arthur L. Carpenter	Although it has been possible to place macro variables in titles and footnotes for quite awhile, Releases 6.07 of the SAS" System and later also include the ability to place BY variable values and names in the titles and footnotes as wen. These options are only documented in SAS Technical Report P-222 and are therefore under utilized. The names of BY variables and the BY variable values are inserted into titles and footnotes using the #BYV AR and #BYV AL options. There are two forms of these options and they will accommodate more than one BY variable. Several examples will be presented using these options.	proc print, by group processing, proc step	source	3
15	California Income Distribution: Using SAS Software to Estimate County Level GINI Coefficients	Daric Brummett	Market research routinely utilizes demographic data to anticipate the consequences of strategic business decisions. One widely used demographic statistic is personal or household disposable income, which indicates purchasing power and, together with other demographic statistics, predicts market performance. Unfortunately, income levels alone many not adequately capture underlying income distributions. Coupled with income level data, the GOO coefficient may significantly increase the accuracy of market research model specification and bolster implied conclusions. This paper defines the GOO coefficient, details estimation methods using SAS software, and applies the method to California county level household data .	regression analysis, proc reg, proc step	source	5
16	Calling External Modules from the SAS System	Richard D. Langston	The new MODULE routines allow for invoking external . subroutines written in COBOL, C. assembler, and other languages within DATA steps, `SAS/IML` steps, and SCL steps. This paper introduces the reader to the MODULE routines. These routines are an outgrowth of the COBOLINT routine that has been discussed in previous articles.	character variable, proc step	source	4
17	Career Insurance for the 21st Century	Arnold Goodman	Is there career insurance for the 21st century? If the 21st century becomes the century for statistics, will it also be the century for statisticians? Likely not. The world is driven by chance and by data, but not by us. Our last four ASA presidents have voiced their concerns. Statistics has grown much more than have statisticians since Wilfrid Dixon introduced BMD software in 1961. Bureau of Labor Statistics projects other occupations to grow much more than statisticians from now until 2005. Where are the technologies of databases, networks and software engineering driving statistics in the 21st century? Where are the economics of markets and pressures of employment forcing statisticians in the 21st century? Have we not lost many customers and their data, and are we not in the process of also losing the analysis? What insurance might attitude and action provide? Does not a species often find it necessary to adopt or migrate, in order to survive the changing environment?		source	3
18	CATALOGER: An Application Development Tool to View, 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
19	Confirmatory Factor Analysis Using PROC CALIS	Diana D. Suhr, James Keaton, Lynne Kelly, Cynthia Finch	The presentation will give an overview of confirmatory factor analysis using structural equation modeling techniques, describe the statistics, show `PROC CALIS` programming code, and discuss the results of the analysis. The results of a reexamination and reanalysis of the Reticence Scale (Kelly, Keaten, Begnal, 1992) will be used as an example of confirmatory factor analysis. Reticence occurs "when people avoid communication because they believe they will lose more by talking than by remaining silent" (Phillips, 1984, p.52). The Reticence Scale, a 24-item self report inventory, was developed to measure reticence using a six-point Likert-type scale of strongly agree to strongly disagree. The Reticence Scale was administered to 269 undergraduate students at Penn State over a 4 year period. No previous study has completed a confirmatory factor analytical procedure to determine the usefulness and meaningfulness of the factors.	proc calis, covariance, latent variable, factor analysis, confirmatory factor analysis, estimate, correlation, proc step	source	5
20	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 times, and varying degrees of detail. Should the developer create multiple 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 they be overdone? Subsets and summarized SAS data sets? All these questions and more will be answered in the pages to follow as the reader learns how to construct a data warehouse.	libname statement, dataset, raw data, data warehouse, cpu time, flat file, macro variable, proc means, proc step	source	6
21	Create a Flexible Parameter Driven Reporting Program By Using Global Variables and PROC REPORT	Karen Dudley	By uSing global variables as parameters, a program can be created which will produce an infinite number of information variations from a single program. The parameters can be toggled on and off by setting up a test flag variable. The parameters may be a single value, a range of values, or a combination of values. This approach requires only a single entry of the parameters for multiple usage. The global variables can be used to subset the data, create flexible `MERGE` statements, and generate meaningful report titles. Additionally, the values can be held in a permanent file to be used by other related programs. Programs developed using this flexible structure can easily be used by bUSiness analysts and other non programming users to create reports to meet their particular business requirements.	proc step, proc format, merge statement, proc report, macro variable, if-then-else	source	4
22	Creating Linked Web Pages for SAS Format Documentation	Jack Hamilton	This paper demonstrates a method of creating Web based documentation for SAS formats, showing differ ent levels of detail (formats vs. informats, numeric vs. character, format library name) on separate pages, with links between them. This method could easily be used for other kinds of hierarchies, such as region / state / county data. Keywords Documentation, FILEVAR= Option on the FILE state ment, Formats, Hierarchical Data, Internet, Links, Web,	proc format, proc step	source	4
23	Creating Shortcuts to Online SAS Help	Jodie M. Gilmore	Under Windows, you can use the SAS System's WINHELP function to create shortcuts to sections of the online SAS System help. For example, you can jump right to the How to ... contents, bring up the Search dialog box, or even jump to an individual topic. This paper gives several examples of using the W1NHELP function.	macro function, sas function	source	2
24	Critical Success Factors in Software Development	Arnold Goodman		management reporting, likelihood function	source	2
25	Cross-Validation Estimates of the Prediction Error in a Random-Coefficients Mixed Model Used to Estimate Emissions from On-Road Motor Vehicles using SAS	Steven Brisby	This paper presents results of a study using Cross Validation to estimate the prediction error associated with several candidate models for vehicle emissions. The basic models are mixed linear models having both random and fixed components. SAS macros were written to generate estimates of the prediction error for each candidate model, thus allowing the analyst to choose the model that has the lowest prediction error. The analysis presented here falls under the classifications of Random Coefficients Models. The SAS `MIXED` Procedure was used to generate restricted maximum likelihood estimates. Performance figures are given for various SAS platforms along with performance tips.	residual, data cleaning, regression analysis, predictor variable, proc mixed, cpu time, response variable, standard deviation, proc reg, stepwise selection, proc step, likelihood function, macro definition	source	5
26	Data Collection: An Internet Based Approach	Aradhana Gokul Janga	The World Wide Web in essence facilitates the communication between computers spread across the world and so networks the people using them. This interconnection provides the ability to exchange information between sources at different locations. In a research environment, dealing with huge amounts of data, this could be a very efficient and cost-effective way of obtaining valuable data and disseminating information. SAS tools could be integrated to realize these uses. This presentation focuses on integrating SAS with Internet related technology for the purposes of collecting up-to-date, error-free, logically correct and meaningful data from the medical facilities. Also these facilities could be provided with information that could prove useful to them, creating an easy-to-use two way information highway. The Need There exist a lot of ways to enter data for purposes of research, analyses and reports.	dataset, merge statement	source	6
27	Data Libraries on Tape	Curtis A. Smith	Big data forces the SAS System developer and user to tum to 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. Wrthin 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 units rather than random access devices, as are disks. What problems does sequential storage bring? Can mu~iple 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	source	4
28	Desing of the Clinical Data Portion of a CANDA to Facilitate Review by CDER	Kay Obenshain	This paper presents the design of the clinical data portion of a CANDA that was submitted to the Division of Metabolic and Endocrine Drug Products at the FDA's Center for Drug Evaluation and Research (CDER) in 1995. `SAS/PH`-Clinical software Version 2.0 (PHC) was customized to provide an interface to data presented in the integrated summary of efficacy and safety (ISE/ISS) sections of the New Drug Application (NDA). The focus of this paper is on the technical aspects of designing the PHC portion of the CANDA, referred to here as the current CANDA.	dataset, laboratory data, vital signs, proc means, clinical data, clinical trial	source	5
29	Developing a PC SAS World Wide Web Database System	Faith Reneé Sloan	The first few months of 1996 have seen an explosion of new Web-related technologies hitting the commercial marketplace as well as being offered freely at numerous sites on the Web. This is especially true in the world of low-cost Web servers, databases, and Web-database gateway software for desktop/personal computers. 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. This paper presents and describes in detail such a system: Developing Educators Resource World Wide Web (WWW) Database System.	dataset, sample	source	6
30	Developing Cross-Platform IntraNet Applications on UNIX	M. Kumar Sagar	This paper illustrates two intranet applications developed using Netscape 3 browser on the client-end and an Apache http server running on a SunOS 5.5 server, at a major pharmaceutical company. The first application is executed from the browser interface, after the user fills in the required input fields. It invokes a PERL cgi-bin script that processes the inputs and in turn invokes an EXPECT script(s). This EXPECT script opens FTP and RLOGIN connections to the remote VAX and ALPHA (AXP) nodes. SAS programs are then executed on the remote VAXlAXP nodes and results are brought back to the UNIX, processed by the PERL cgi-bin script, and presented to the user in the HTML format. A Second application exploits the strengths of the SAS System by employing SASlConnect to make a remote connection to the VAXlAXP nodes and transfer the SAS datasets to the UNIX node. This is again done from within the Netscape 3 browser on the client-end where the user fills in the required input fields.	dataset, error message, view, macro definition, create view, proc sql	source	9
31	Diagnostic Plots in Regression Analysis	George C. J. Fernandez	In Multiple linear regression models, problems arise when serious multicollinearity or influential outliers are present in the data. Failure to include significant quadratic or cross-product terms result in model specification error. Simple scatter plots are most of the time not effective in revealing the complex relationships of predictor variables or data problems in multiple linear regression. However, partial regression plots are considered useful in detecting influential observations and mUltiple outliers; partial residual plots or the added-variable or component-plus-residual plots are useful in detecting non linearity and model specification errors. The leverage plots available in SAS JMP software are considered effective in detecting multicollinearity and outliers. The VIF-plot, which is very effective in detecting multicollinearity, can be obtained by overlaying both partial regression and partial residual plots with a common centered X-axis.	regression analysis, proc reg, predictor variable, multiple regression, outlier, scatter plot, response variable, macro definition, influential observation, proc step	source	8
32	Economic Development and Women's Social Status: A SAS Analysis	Seid M. Zekavat	The pUIpose of this study is to illustrate the use of several of the SAS System multivariate procedures. Specifically, the study shows how `SAS/STAT` and SASIETS procedures are used to estimate the relationship between the social status of women and numerous indicator variables such as per capita Real Gross Domestic Product, the Human Development Index (HOI) and the Gender Development Index (GDI). Preliminary results indicate that, unless there are cultural and religious barriers, women in developed countries enjoy a higher social status than do those in developing countries. `PROC CORR`, `PROC REG`, `PROC` STEPWISE and `PROC` RSQUARE are some of the SAS procedures used in this paper.	regression analysis, data science, r-squared, proc step, response variable	source	7
33	Efficiency Ideas for Large Files	J. Meimei Ma, Andrew H. Karp	This tutorial presents options you should consider implementing when using the SAS System for handling large files. We touch on a variety of topics and approaches, as opposed to dealing with only a few in depth. Using compression and index files are among the storage techniques covered, along with ways to plan your programming strategy to incorporate appropriate efficiency concepts. The strategies and hints discussed are directed toward programmers, data managers, project managers, and anyone else interested in cost effective solutions to SAS System programming problems.	dataset, proc step, proc sql, numeric variable, sas programming	source	9
34	Estimation of Shelf-Life and the Minimum Effective Dose Using PROC MIXED	Louis Munyakazi	The concepts of shelf life and the minimun effective dose estimation have been studied by numerous authors. Some of the recommended approaches include the simple regression problem of computing the confidence interval around the regression line and the idea of non overlapping intervals. We extend the same ideas to include different covariance structures likely to represent the true functional relationship of the responses over time. The elements needed are easily extracted from the new procedure `PROC MIXED` of `SAS/STA`~.	covariance, covariance structure, confidence interval, proc mixed, proc step, regression analysis	source	6
35	Exploiting SAS Software Using JAVA Technology	Donald J. Henderson, Barbara Walters	This paper describes how to use Java'M technology with SAS software. SAS Institute currently offers several Java development tools that allow applications to access SAS data and take advantage of SAS computational capabilities. This paper describes how to use these class libraries and address clienllserver configuration and performance issues.		source	5
36	Exploiting SAS to Work with Other Software: Paradox Database Programs to Write SAS Code and WordPerfect Macros to Clean Up SAS Output	Cecilie Birmer, Kathleen Martin, Gary Sulyer	Data cleaning in SAS and getting print quality tables from `PROC TABULATE` in SAS can be very time consuming and tedious. In my presentation, I will show how we used Paradox (a database program) macros to output data cleaning code. I will also illustrate how we used WordPerfect macros to generate print-quality tables automatically from SAS output. Allowing other programs to work with SAS can save hours of programming, improve efficiency, and allow non-SAS users to assist substantially in SAS programming.	proc step, proc tabulate, proc printto, data cleaning, data step, macro definition	source	6
37	Exploring Ways to Compare Eight Diverse Hospitals	Anthony M. Dymond	A significant and general problem in data exploration is to decide how to use a set of variables to compare a somewhat diverse group of subjects. Specifically, how does one find appropriate dimensions for comparison and then calculate comparative statistics on these dimensions. This problem is illustrated here by a comparison of eight rather different hospitals based on five variables measuring their patient discharge characteristics. Multidimensional scaling is used to let the data suggest three dimensions for comparison. Interpretation of these three dimensions is improved by axes rotations, and a comparative score on each dimension for each hospital is calculated. Plots of the hospitals in these dimensions then provide an easily interpreted comparison. Although Illustrated here using hospital data, this process is generally applicable to a variety of situations.	factor analysis, correlation, proc step, euclidean distance	source	6
38	Factors Influencing Life Expectancy: Using SAS Regression Procedures in a Multi-Country Study	Kamal Shoukry	This study considers some of the influential factors which are believed to affect life expectancy according to the United Nations' institutions. Those factors and their symbols are: life expectancy at birth (LFXP), health services (HSVS), safe water (SFWR), sanitation (SANT), % daily calorie supply per capita (CALCP), % adult literacy rate (LTRC), % combined 1st-2nd-3rd levels enrollment ratio (ENRR), daily newspapers copies per 100 people (NEWP), televisions per 100 people (TVNU), real GOP per capita (RGDP), and GNP per capita in US$ (GNPC). The data used here covers 127 countries classified by the United Nations into three categories high, medium and low according to their human development.	predictor variable, correlation, regression analysis, multiple regression, proc step	source	5
39	Flying Between MS-Excel and SAS Data Sets	Su-Ting Luo, Edward Wei-Tarng Lee	This paper introduces the `SAS/ACCESS` interface between Microsoft-`EXCEL` worksheets and SAS data sets. Using `PROC ACCESS` or `PROC DBLOAD` to fly back and forth between PC files fonnats, one can view, access, convert, load, select, subset or create data sets in both terrain. The author will explain several hands-on tips in SAS syntax or options. Flyer-beware limitations that may trap the format or contents of the data sets will also be discussed.	excel, dataset, proc step	source	4
40	How SAS Users Can Benefit from the Internet	Helen D. Carey, Ginger Carey	TIle Internet is the largest collection of interconnected networks in the world. As the Internet grows exponentially, so does the number of SAS users accessing it as part of their daily work activities. TIle vast amount of information on the Internet can be overwhelming. This paper describes the access tools of the Internet, the location of helpful SAS resources, reasons for using the Internet, and how to find information so that you benefit from the Internet and not just access it You can enjoy an enormous amount of information on the Internet	isp, service provider	source	10
41	How to Use the SAS ODBC Driver to Mine Your Data Warehouse via the World Wide Web	Marvyn M. Lindsey, Charles E. Shipp	The objective of this paper is to discuss various methods to access SAS data from the World Wide Web. The' use of tools such as the SAS ODBC driver and the SASlIntrNet components to mine your data warehouse will be discussed. How to set up and to implement access to SAS data sources using the MS Internet Information Server 3.0 via Windows NT Server 4.0 will be highlighted.	proc sql, dataset	source	6
42	How to Use the SAS Software to Evaluate Screening Tests Using Predictive Values in Conjunction with ROC Curves	Richard Severino	Development of new screening tests is ongoing in the medical and biotechnology fields, be it for a medical condition or in an attempt to predict outcome. The screen usually involves a measurement that is then compared to a cutoff point in order to classify the test subject into one of two categories. In developing screening tools, researchers often concentrate on the sensitivity and specificity of the test, and use ROC curves to evaluate the discriminating power of the screening test often ignoring the predictive values of the test. The predictive values of the screen are equally important as they measure the accuracy of the prediction made on a subject whose true condition is unknown. By examining the predictive values, a cutoff point may be chosen such that it minimizes the false positive and false negative rates. A method for obtaining and examining these values using SAS is presented.	roc curve, logistic regression, proc step, logistic distribution, dataset, regression analysis, sample, type i error, proc gplot	source	7
43	Implementation of Benefit Capitation Algorithm for Managed Health Care	Zhirong Mao, Yuejing XU	In this paper several benefit capitation algorithms are implemented with SAS codes to meet the dynamic needs from health care industry. Stop-loss points are detennined for total number of treatment sessions and total treatment cost. Two calculation approaches are presented with managed care examples in deductible-copay-capitation benefit paradigms.	proc step, proc sort	source	5
44	Incrementing Year Fields in Flat Files to Test Year 2000 Compliance	Brian Bouvier	One critical task to ensure correct handling of dates .in the 21" century is to parallel test SAS" programs that read flat files. In an environment with hundreds of user-written SAS programs each reading a different extract of a master file, ensuring proper parallel testing is a daunting task. This MVSutility allows end-users to "age" their own data wHhout requiring knowledge of JCL or the SAS system and is customizable for use with any flat file. Once the year fields have been identified in the flat file, the utilHy increments all the year fields by a user-supplied amount and produces a new fde. This enables the end-user to produce as many versions of the extract as required to completely test their programs for years 1997,1999,2000,2001, and beyond.	insurer, flat file, macro definition	source	5
45	Introducing Release 6.12 of the SAS System for Personal Computers	Russ Robins	Release 6.12 of the SAS System for Per sonal Computers provides support for Mi crosoft Windows (version 3.1 using Win32s technology, Windows 95, Windows NT), and OS/~ (Warp). With this release, the SAS System incrementally improves both its user interface and its interoperability with other PC applications. This paper discusses both new features and incremental enhance ments to the user interface. This paper also describes the SAS System for Windows 95 and Windows NT 4.0, detailing the new user interface features that are consistent with the look and feel for Windows 95 and Win dows NT 4.0. The SAS System Viewer, pro duction for 6.12 is described, as is the GUI based SETUP program, which first ap peared in 6.11. Interoperability topics in clude Windows Explorer integration, elec tronic mail, Lotus Notes, OLE, OBDC, and access to external DLLs. This paper also discusses the support provided by the SAS System for Windows NT.	dataset, put statement, sas base	source	9
46	Investigating Gender Empowerment: A SAS/STAT Application	J. Richard Earley	Does gender empowerment yield positive rewards to women? This paper answers that question by investigating the relationship between gender empowerment and various outcome measures such as longevity, the earned income share of women, the percentage of administration and managerial jobs held by women and the number of seats women hold in their country's parliaments. Using a data set of 174 countries from the United Nations' Human Development Report for 1996 (longevity data for 1993), the paper illustrates how to estimate and test a regression model developed using `PROC REG`. The paper also shows how to perform regression diagnostic testing using the SAS System.	response variable, standard deviation, anova, proc step, regression analysis, proc reg	source	4
47	Logistic Regression Modeling in Complex Surveys: Examples Using LOGITSE SAS Macro	Jesse Albert Canchola, Brian D. Marx, Joseph A. Catania	Traditional formulae for standard errors and subsequent statistical significance tests implemented in various popular statistical packages are based on the premise that the data are a simple random sample (SRS) of observations from a super-population. Equivalently, the observations are assumed to be independent and identically distributed (110). For complex analytic surveys, these assumptions are almost always invalid leading to potentially incorrect inferences due to the failure to adjust relevant standard errors of the parameters. We give examples of binary type responses which are frequently encountered in survey research where a logistic regression analysiS would be meaningful, and use the SAS macro, LOGITSE, to take the c/uster-correlated nature of the complex survey design into account.	logistic regression, logistic distribution, proc calis, regression analysis, design effect, odds ratio, type i error, sample, macro call, significance level, response variable, predictor variable, correlation, random sampling, estimate, null hypothesis, cv, dataset, covariance, likelihood function, outlier, main effect, proc step, confidence interval	source	10
48	Managing Large Health Insurance Claims-Based Files: Tricks and Techniques	Miriam G. Cisternas, Marcia Stoddard	Administrative databases are increasingly being used for research purposes. In health outcomes consulting, we often use insurance claims data to profile disease-specific treatments and costs. The advantages in using claims data over primary data are that they are cheaper and faster to obtain. The disadvantages are that they can be difficult to use and sometimes have severe limitations relative to the research questions. We would like to share some of our experiences extracting and managing health insurance claims to answer pharmacoeconomic questions. This paper is divided into two parts. We first discuss methods for extracting data to create a 'selection" file. Such	insurance claims, health insurance, insurer	source	4
49	Methods of Finding a Small Group of Records in Two Million Without Using Merge	Lynn G. Palmer	Searching for particular records in files or databases is easy with SAS software, unless you are trying to find 1000 records in among 2 million. When using large files or databases, there are methods that can be applied to prevent space errors from occurring. This paper will present JCL, SQL, and SAS	proc sql, macro facility, proc step, macro variable, filename statement, macro quoting function	source	6
50	Monitoring a Cohort of Blood Donors and Recipients Using SAS/AF Frame	Monique Bryher	The Transfusion-Transmitted Virus Study (TTVS) was a National Heart Lung Blood Institute (NHLBI) sponsored investigation during the 1970s of characteristics of blood donors and recipients in four geographic areas of the United States. Its immediate	dataset, proc step	source	6
51	Monte Carlo Simulation Approach to Assess Health Care Claims Risk	Zhirong Mao	This paper that utilizes statistical functions of the SAS system presents a simulation approach to approximate the probability distribution of Claims, which is determined jointly by several random variables, such as health care access, average length of stay and average cost per unit. The simulated distribution provides an effective tool to assess the claims risk exposure without deriving the claims probability distribution in analytical form. Use of this approach is extensible to risk management, insurance product pricing and provider network development.	poisson distribution, normal distribution, lognormal distribution, proc freq	source	4
52	On Time and Under Budget: Guidelines for Success Using the SAS System with Traditional Structured Techniques	Marc Daniels, Clark Roberts	A few years ago, Community Care Network, a provider of services to the health care industry, faced a major crisis. One of the services they provided to trading partners was initially provided on an ad-hoc basis. As business increased, CCN staff found it difficult to meet schedule expectations and were in jeopardy of losing business. The Information Services group recommended a system that would provide CCN with a uniform approach to providing their trading partners with this product. Because of the complexity of the process traditional structured approaches were used in all phases of the project, and SAS System was chosen over other alternatives as the development environment. The final product was a bug free system that was produced on time and under budget. In reviewing the various stages of the EDIP project we will point out the critical factors we believe contributed to it's successful implementation.	outlier, dataset	source	4
53	Output Tabluated Information into Data Sets	Zhirong Mao, Yuejing XU	In this paper several methods using `DATA STEP` operations and procedures are discussed to simulate the output from `PROCEDURE TABULATE`. Tabulated information is then saved into SAS data sets for further processing. These methods overcome the limitation that `TABULATE` procedure doe not output report into data sets. Examples show advantages and disadvantages of these methods.	proc step, proc report, proc means, proc tabulate, proc print	source	4
54	Population Adjusted Kaplan-Meier Curves	Jimmy Thomas Efrid	In this paper we present a statistical method for comparing data from two populations that may differ substantially with respect to outcome related covariates. Typically, this method involves comparing a study population with that of a reference population and using variables in the reference population to adjust or standardize the data in the study database. The databases are compared with respect to the time from baseline to a particular event.	sample, logistic distribution, confidence interval, regression analysis, proc calis	source	6
55	Presentation of Simple Statistics Using Multiple By-Variables	Kathryn A. Valdes		macro variable, excel, proc freq, proc step, dataset, label statement, macro definition	source	6
56	Reliability Data Analysis in the SAS System	Donna O. Fulenwider, Gordon Johnson	The `RELIABILITY` procedure, a recent addition to `SAS/QC` software, provides tools for reliability and survival data analysis as well as for recurrence data analysis. You can use this procedure to construct probability plots and fit life distributions with right and interval censored lifetime data, fit regression models, including accelerated life test models, to combinations of right, left, and interval censored data, and analyze recurrence data from repairable systems. These tools benefit reliability engineers and industrial statisticians working with product life data and system repair data. They also aid workers in other fields, such as medical research, pharmaceuticals, social sciences, and business, where survival and recurrence data are analyzed.	confidence interval, regression analysis, right censoring, proc calis, interval censoring, likelihood function, data science, predictor variable, maximum likelihood estimation, estimate, weibull distribution, dataset, proc step, sample, censoring, binomial distribution, logistic distribution	source	10
57	Repeated Measures Analysis on Preservice Teachers' Attitudes Towards the Subjects Taught in School	Leping Liu, George C. J. Fernandez	This paper will present repeated measures analysis on elementary and secondary preservice teachers' attitudes towards the five subjects-mathematics, science.ianguage, social studies and computer technology, at four attitude levels-<mjO)lllent, motivation, importance, freedom from fear of the subjects. Significant.differences were found between elementary and secondaIy preservice teachers attitudes towards science, social studies, and computer technology. For all the five subjects, significant differences were found among the four attitude levels. For science. social studies, and computer technology, interactions were found.	repeated measures, repeated measures anova	source	6
58	Retrieving Weather Data from the Internet and Analyzing Using the SAS System	Richard D. Langston	This poster attempts to describe !WO distinct processes: 1) obtaining weather data from Internet sites and 2) analyzing the data using the SAS" System. There are !wo different Internet sites that are described: 1) the National Climatic Data Center OASIS site, and 2) the hurricane data repository at Purdue University.	proc step, dataset	source	6
59	SAS/SPECTRAVIEW Software and Data Mining: A Case Study	Stuart Nisbet, Ronald E. Stogner, Aaron Hill	Advances in information technology are allowing organizations to collect and store larger amounts of detailed data. Today's challenge is for companies to tum this information into a competitive advantage. Application areas, such as data mining, database marketing and decision support, hold the promise of tuming this information info profits. One key component to this process is data visualization. This paper presents a practical example of using data visualization technology to assist companies in the decision support process. `SAS/SPECTRAVIEWSOFTWARE` is used to spot key relationships and trends in a large data set. `SAS/SPECTRA` VIEW Software and Data Mining Major companies today are faced with a growing dUemma. They have amassed large quantities of data related to their businesses and their customers. The challenge for these companies is to tum this data into profitable information.	data mining, predictor variable, sample, bar chart, response variable	source	5
60	SAS/Warehouse Adminstrator Usage and Enhancements	Tony Fisher, Terry L. Lewis	SASlWarehouse Adminlstrator software makes it easier to build. maintain. and access data warehouses by bringing the strengths of the SAS System relative to data warehousing together into one consIStent interface and supplying additional data warehouse oriented capabilities .not currently available in other SAS system products. This paper explains how to use SASlWarehouse Administrator and describes many of the enhancements that have been added to the first release of SASlWarehouse Administrator.	data mart, data warehouse, sample, dataset, proc step	source	9
61	Smokin' with UNIX Pipes	Kimberly J. LeBouton, Thomas W. Rice	The `FILENAME` PIPE option can have your SAS programs "smokin" in no time. `FILENAME` PIPE allows you to read in the output of any UNIX operating system command into a SAS dataset. As a SAS dataset, you now have the flexibility of the SAS language, which will also cut down on the amount of UNIX script writing needed. Managing Data with Is Command Three examples will be provided to demonstrate the use of `FILENAME` PIPE. The first example retrieves the output from the UNIX Is command. The Is command displays a listing of dataset names. Output 1 is a partial list from the command Is -d /appVicpl*. The-d option shows only the directories, not the individual files associated with each directory. Figure 1 shows the DATA `_NULL_` step that reads in the dataset names (dsn) and then stores the names in macro variables, named in1 to inn. The small macro program, MUBNAME, uses these macro variables to genemte the proper `LIBNAME` statements.	dataset, proc step, macro variable, proc sql	source	4
62	Some Applications of Data Mining Tools in Database Marketing	Jiamin Liu	This paper presents some results of an empirical study on database marketing using CART, artificial intelligence-based algorithm (AI) and Legit model. The robustness of the three different approaches is evaluated using a large dataset.	logistic regression, proc hpsplit, regression analysis, test set, logistic distribution	source	6
63	Submitting Batch SAS Jobs under Windows 3.1 and Windows 95	Jodie M. Gilmore	This paper illustrates several methods of submitting batch SAS jobs in the Windows 3.1 and Windows 95 environments (Windows NT users can use these methods as well). Windows 3.1 users leam how to edit the REGEDIT database so that double-clicking on a SAS file icon submits the program in batch mode. Both Windows 3.1 and Windows 95 users benefit from the discussion of other methods, such as running a job from the DOS prompt, adding a second "batch-only' icon to the SAS System program group, and dragging and dropping file icons onto the SAS.EXE icon. The paper also includes a brief discusSion of using a few system options that are useful in batch mode, as well as instructions on how to submitting batch jobs from display manager and submitting more than one batch SAS job at a time.	batch job, options statement, sas batch, sas output, sas log, proc printto, sample, proc step	source	4
64	Survival Strategies and Solutions for the new Millenium - Year 2000	Kirk Paul Lafler, Sunil Kumar Gupta, Charles E. Shipp	As the tum of the century approaches, critical challenges face organizations world-wide. Will mission-critical computer applications be equipped to handle the numerous calculations and comparisons required by businesses to successfully operate? And how can organizations smoothly transition into the next century without impacting current operations and staffmg levels? If estimates are accurate, as much as 90 percent of all business applications will be impacted in some way by the coming millennium. A successful strategy is needed to ensure cost-effective, timely, and lasting solutions.	sysdate9, date variable, proc step	source	6
65	Taking Advantage of the SAS System on Windows NT	Mark W. Cates	Windows NT is fast becoming the universal desktop client operating system as well as an important file and compute server for mission critical applications. This paper presents a discnssion of the state of Windows NT and how the SAS System Release 6.12 for Windows takes advantage and integrates with the operating system. Areas such as the user interface, OLE and Web integration are presented. Data access and Microsoft BackOffice integration, and hardware considerations are also presented.	proc sql, sql server	source	10
66	Taking the Mystery Out of SAS Macro When Using CALL SYMPUT	K. Larry Landers, Monique Bryher	The SAS manuals and guides provide good information about using CALL SYMPUT to write data values to the macro environment, and, then, to retrieve those values in subsequent DATA steps or PROCs. Even so, it is easy to write seemingly valid code, only to find that the results are not as expected; or to use the same code in slightly different contexts, only to find that one works, while the other does not. Much of the confusion in such situations can be attributed to an insufficient understanding of what SAS is doing in the SYMPUT routine. Through a series of examples using CALL SYMPUT statements, we will illustrate how SAS processes macro variables and discuss the considerations for handling character, numeric and null values within, or between, DATA steps.	macro variable, call symput, sas base, put statement, missing value, input statement, numeric variable, sas log, macro definition, character variable	source	6
67	Ten Great Reasons to Learn SAS Software's SQL Procedure	Kirk Paul Lafler	The `SQL` Procedure has so many great features for both end-users and programmers. It's fun, easy to learn and use, and can often result in fewer and shorter lines of code than using conventional DATA and `PROC STEP`	proc sql, proc step, select, where statement, group by, order by, alter table, delete, proc contents	source	6
68	Testing Homogeneity of Variance	Shuqiang Zhang	Homogeneity of variance is a major assumption underlying the validity of many parametric tests such as the t-test and ANOV A. It also represents the nuU hypothesis in substantive studies that focus on cross- or within-group fluctuation in variability. This paper explains how 12 homogeneity of variance tests, many of which have no corresponding SAS software options, can be conducted using `SAS/STA` r.	proc step, sample, correlation, proc glm, degrees of freedom, proc genmod, anova, null hypothesis, significance level, repeated measures, sample size	source	4
69	The Problem with NODUPS	Jack Hamilton	The NODUPLICATES option in the `SORT` procedure supposedly deletes duplicate records from a SAS data set. In reality, its actions are somewhat unpre dictable, due to an oddity in `PROC SORT`. This oddity can also lead to vastly inflated processing time in any `PROC SORT` that uses a DROP or KEEP statement. This paper presents examples and suggests some al ternatives. Keywords Efficiency, NODUPKEYS, NODUPLICATES, `PROC SORT`, `SORT` Procedure, SQL	proc sort, proc step, merge statement, proc sql	source	4
70	The SAS Multidimensional Database and Its Uses in Application Development	Bill Powers	The SAS multidimensional database (MDDB) provides an efficient and powerful data storage structure for summarized data. Data stored in the MDDB data structure can be used by standard SAS procedures and in applications developed with the SASIEIS and `SAS/AF`'" modules. This tutorial discusses when and how to create an MDDB both with programming statements and through the SASIEIS point-and-click interface. Consideration will be given to efficiency issues and updating an existing MDDB. Examples will be presented demonstrating how to use an MDDB with SAS procedures, with selected SASIEIS objects, and in `SAS/AF` FRAME entries.	class variable, dataset, proc step, raw data, proc means	source	5
71	The SAS/IntrNet Dispatcher	Donald J. Henderson	The Application Dispatcher component ofSASlIntrNet Software allows a user to run a specified SAS Application program on demand and return the generated Internet Content to a Web Browser. The Application Dispatcher is one of the compute services of SASlIntrNet software and is composed of the following components: 1. The Application Broker CGI The link between the Web Browser and the compute services. Resides on the Web Server and forwards requests for processing to the Application Server(s). 2. The Application Server An SCL program that listens for requests from the Broker and runs the specified SAS Application. 3. SAS Applications SAS programs that perform specified processing. They generate the results (Internet Content) which are sent back to the Web Browser by the Application Server via the Dispatcher. The MDDB Report Viewer is an application that allows users to access and drill down through an MDDB data store.	macro variable, macro definition, sample	source	9
72	The Select Statement	Jack Hamilton	The SELECT statement is an alternative to nested IF statements. It is often easier to code and understand (especially six months later) than the equivalent IF based code. This paper presents the syntax and sev eral typical uses of the SELECT statement. Keywords IF Statement, IN= Dataset option, `MERGE` Statement, SELECT Statement Syntax of the SELECT Statement The SELECT statement is described in Chapter 8, SAS Language Statements, of the SAS Language Refer ence. It has two variations, with and without a select expression. SELECT with a select-expression Use this form of the SELECT statement when you want to base your action on the value of a Single variable or expression. SELECT (select-expression); WHEN (expressionl, etc.) statement-l; WHEN (expression-n, etc.) statement-2; [lots more WHEN clauses if you want]	select, if-then-else	source	4
73	The Social Science Database Archive (SSDBA) Moves to the World Wide Web	Stuart E. Wugalter, Jeannie Lum, Colin Campbell, Alan Wang, Desemona Carddoza	This paper will present the reconstruction of an academic data archive using a combination of the SAS Data Warehouse model (see, for example, SAS Institute, 1995), SASlIntrNet software, and other SAS software. The Social Science Data Archive (SSDBA), located at California State University Los Angeles (CSLA), provides public data to faculty and staff for research and instructional purposes. After 5 years in existence, both the hardware and the interface needed to be updated. In addition, it was determined that a data warehouse model would be used to improve the administration of the SSDBA. There are many SAS system examples of the use of a data warehouse model for both information technology departments and businesses, but not for an academic data warehouse. This paper will explain the process of migrating from a character to a web based interface using SAS software.	dataset, data warehouse	source	6
74	The Sums of Squares Are All the Same - How Can the LSMEANS Be So Different?	Lori Potter, David J. Pasta	In analysis of variance (ANOVA) and analysis of covariance (ANCOV A) using `PROC GLM` in `SAS/STAR` Software, specifying the least squares means (LSMEANS) adjusts estimated effects for the other factors included in the model. Variables that take on only two values (binary variables) can be treated as categorical variables (and therefore included in the CLASS statement), or as continuous variables (and thus excluded from the CLASS statement). This choice can affect the least-squares means, sometimes dramatically, especially if the covariate is extremely imbalanced. The least squares means are also affected by whether or not the OBSMARGINS option is requested. In this paper, the authors examine the least squares means in a cost of treatment model for mental health inpatients, and they demonstrate how the specification of equivalent models in `PROC GLM` can affect the outcome generated from LSMEANS.	proc glm, proc genmod, proc step, estimate, response variable, binary variable, treatment effect, anova	source	6
75	The Use of Central Analysis File or Patient File in the Analysis and Management of Clinical Trials Data	Ajit Gemunu de Silva	Data used in the analysis of clinical trials are usually organized along several functional data set types. These data sets capture infonnation collected with an industry survey instrument called the Case Report Fonn. Different data sets may have different numbers of observations per patient, depending on the frequency with which pieces of patient infonnation are required. One central patient data set is often created to ensure the correct number of observations are used in different analyses and to ensure correct subsetting by treatment group (placebo versus dosed) and type of analyses (safety versus efficacy). The analysis data set or patient file is used for these subsets and contains flags indicating to which group a patient may belong. In addition, other computed variables derived from, but not originally contained in, the raw data are also included.	clinical trial, raw data, data science, proc freq, adverse events	source	3
76	Using SAS as a Clinical Data Repository	Carl R. Haske	Managing clinical trials data requires the efforts of a large team of people, including clinical data managers, data entry personnel, programmers, and statisticians. A system that enables concur rent access to data at several stages of the database life cycle enhances the overall work process and accelerates database development. SAS is the software package of choice in the phar maceutica industry for analyzing clinical data and interpreting results. Given that SAS is generally used as a front-end tool to produce interpretive results, it makes sense to consider SAS based systems for use on the back-end for managing data. STATPROBE, Inc., has developed an integrated data manage ment system that incorporates security and tools to assist in the valid production and analysis of the clinical data.	clinical data, dataset, audit trail, clinical database, data cleaning, data governance, clinical trial, proc step	source	6
77	Using SAS Multivariate Procedures to Estimate Human Development Indicators: The Case of Education and Human Development	Joseph E. Earley	This paper illustrates how the BAS System may be used in investigating the relationship between human development and education. Numerous BAS procedures are used on a data set developed by the United Nations Development Programme for 130 countries. In particular, `PROC REG`, `PROC` STEPWISE, `PROC` RSQUARE and `PROC CORR` are used to estimate the relationship between human development and numerous education indicator variables such as public expenditure on education as a percent of GNP.	proc step, regression analysis, response variable, proc reg, anova, estimate	source	4
78	Using the Data Warehouse Model is Streamline and Accelerate New Drug and Medical Device Development	Martin J. Rosenberg	Many companies that developed their own clinical data management systems in the 80's and early 90's settled on a model that enters and stores data in an On Line Transaction Processing (OL TP) oriented Relational Data Base Management System (RDBMS) and then moves that data to the SAS~ System for analysis. This choice was made for a number of reasons: many information technology (IT) professionals did not consider SAS software to be a true relational database; companies selected their clinical data management systems at a time when the SAS system lacked certain relevant features found in traditional RDBMS systems; MIS departments involved in the selection of a database tended to chose the tools with which they were already familiar. The original purpose of an RDBMS was to manage a company's operational data in the form of individual transactions.	clinical data, clinical data management, data governance, clinical trial, relational database, clinical research, data science, data review	source	5
79	W-O-W Data Warehouse, SAS/OR and the WEB: A Decision Support System (DSS), Using a Data Warehouse Designed for Project Management and Optimized for SAS/OR, Distributed on the WWW	Art Rerecich	Although our operational systems are very adept at processing transactions they don't lend themselves to ad hoc reportage or analysis. Operational and senior management need distilled infonnation from these systems for effective decision making. Eighteen months ago we were tasked to deploy a Decision Support System for executive level program surveillance and On-Line Analytical Processing (OLAP). We designed a Data Warehouse with the SAS system, running in a ClientlServer configuration (UNIX & Windows 95), `SAS/OR`, with output distributed across the World Wide Web (WWW), to accomplish this task.	excel, ods excel	source	6
80	Web Data Distribution - Moving Beyond Static Pages: The Web Empowerment of Decision Makers	Art Rerecich, Deanna Castorena	The users of our Decision Support System require fairly complex infonnation available in several formats. Since the Web is our designated deployment environment we bad to find ways to let our users pull information from our Data Warehouse and use it in a variety of ways, from presentation to analysis. We required interactive pages !bat pulled data based on user input and then created output in multiple fonnals. The input would have to accommodate multiple select criteria (time horizons, variable selections, sort order, etc.). The output would need to support several fonnalS (text reports, `EXCEL`, graphics suitable for PowerPoint presentations, etc.). This paper describes some of our solutions: I. Dynamic and Interactive Web Page Design: a) HfML Fonus, Frames and the "Frame Chain" Data Down Select Approach b) CGI Scripting in PERL c) SAS htmSQL 2.	excel, proc sql, delimited file, macro variable, macro definition	source	7
81	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 (e.g., selecting only those observations which occurred after a particular point in time); 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, datetime variable, date variable, datepart, time variable, numeric variable, dataset, raw data, proc freq, proc step, mdy, datetime, time, if-then-else, subsetting if, time series analysis	source	7
82	Year 2000 Hex Storage	Kirk Paul Lafler, Sunil Kumar Gupta, Charles E. Shipp	Legacy systems that store, display or print dates in two-characte:crepresentation for year (,97, '98, '99, ... ) will have an ambiguity problem when any of the year values go beyond the year 1999. Serious time-span calculations and other problems will then occur. Fourth generation language (4GL) date functions, such as the SAS function YEARCUTOFF provide a user selected lOO-year window and is good for many situations. Another solution is to readand write the storage using hex to n:tain the old record layouL Two hexadecimal bytes can represent year values from 0 to over 65,000, removing the lOO-year IimiL This	flat file, date variable	source	3
83	Year 2000 Solutions	Kirk Paul Lafler, Sunil Kumar Gupta, Charles E. Shipp	SAS Institute has providedan excel1entsysfem/program option. YEARCUTOFF-ccyy. but there are seveml other options for SAS programmers available also. lbis poster discusses the stepS of a successful program. Companies will need to examine the role of their major systems that have dates and select the solution appropriate for each SAS program/system. Steps for a planned transition to the year 2000 are herein presented.		source	2
84	You Want Me to Put HOW MANY Numbers in this Document?	Cynthia L. Williamson	Maybe someone else is responsible for explaining the results in tables, but you are responsible for creating the data that go into the tables. What do you do when you have hundreds of variables and you need to report both unadjusted and adjusted means for the treatment and control group, along with the associated t statistics and p-values? Well, one answer is to use the SAS System (`DATA STEP`, `SAS/STAT`, and the SAS macro facility) to get the numbers and then use Microsoft Word's mail merge facility to create presentation-quality tables.	proc glm, proc step, proc genmod, response variable, ascii, txt, if-then-else, proc print	source	8

Back to Conf go 1998 go 1996

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.