WUSS 1993 Papers - Index

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

go 1994

#	Title ×	Authors ×	Abstract ×	Keywords ×	Extras	Pages
1	180 Terrific Ideas for Using and Learning the SAS System	Helen D. Carey, Ginger Carey	Perhaps you have seen Life Little Instruction Book by H. Jackson Brown, Jr., which has 511 suggestions, observations, and reminders on how to live a happy and rewarding life. Well, it is time for us SAS enthusiasts to have our own list of both new ideas and reminders of how to be more productive and happy. There is a wealth of information available from SAS publications, SAS users, experience and the SAS-L electronic discussion group.	proc step, data step, dataset, character variable, proc print, sample, proc freq	source	6
2	A Macro for Personalized Odds Ratios	Alice Pressman, K. L. Stone, Estie Hudes, J. L. Tao	In this paper, we offer a macro which manipulates output from SASlSTAl" (1) `PROC LOGISTIC` (`PROC` LOGIST in V5.18). Our macro combines the output from `PROC LOGISTIC` with other information to produce a personalized Odds Ratio (OR) and Confidence Interval for each independent predictor. This macro allows you to choose the unit of interest for each independent variable in your model. The macro calculates ORs and Confidence Intervals based on an increase or decrease in the unit of interest, while simultaneously adjusting for any number of covariates. If the standard deviation is the unit of interest, the resulting OR is considered a Standardized Odds Ratio.	logistic distribution, proc step, proc logistic, logistic regression, standard deviation, if-then-else, odds ratio, regression analysis	source	4
3	A Modification of Two-Stage Least Squares in Quasi-Experiments	Alex Ezuzides, Raymond C. Peck	The California Department of Motor Vehicles establishes curriculum guidelines for education courses designed to update the driving skills and knowledge of drivers 55 years of age or older. This program has been evaluated to determine its effectiveness in reducing accident and conviction involvement. The evaluation model represents an experiment with nonrandom assignment to treatment and control groups. There are many factors, including selection bias, that influence both the behavioral outcome and the assignment to treatment, that may not be measured. Application of ordinary least squares techniques in quasi-experiments leads to biased estimates. In this paper, a triangular structural model is used and a modification of the two stage least squares technique (using `PROC SYSLIN`) is proposed in order to correct for heteroscedasticity in the errors. The resulting estimated treatment effect is shown to be consistent.	correlation, proc calis, treatment effect, covariance, proc reg, response variable	source	9
4	A SAS Macro for Power and Sample Size Calculations for Historically-Controlled Studies with Dichotomous Endpoints	Andrew Dagis, Chul W. Ahn	Historical controls are frequently utilized for evaluating relative effectiveness of therapies. Even though a number of software programs are widely available to compute power and sample size for randomized comparative studies, programs to calculate power and sample size for non-randomized studies are not readily available to many statisticians. We have programmed a SAS macro to compute pow~r and sample size for non randomized studies with dichotomous endpoints. The use of the SAS macro is illustrated using actual data in the area of bone marrow transplantation. In this instance, a study is designed to compare a new diagnostic procedure to a data set of 200 patients tested with an established procedure.	sample, sample size, significance level, macro definition	source	6
5	A SAS Macro to Test the Significance of Differences Between Parameter Estimates in PROC CATMOD	Linda R. Ferguson, Mel Widawski, Dawn M. Upchurch	A SAS macro, DIFFPARM, is described that extends the capabilities of the `CATMOD` procedure for analyses of multinomial logistic regression models. The DIFFPARM macro uses the results of PROe CATMOD to calculate significance tests for differences In parameter estimates across categories of the outcome variable. These significance tests enable more elaborate	estimate, response variable, logistic regression, regression analysis, logistic distribution, multinomial logistic regression, predictor variable, proc step, covariance, macro definition	source	5
6	A SAS Program for Sample Size Determination for Equivalence Trials	Akiko Tomita, Chul W. Ahn, Joyce C. Niland	Adequate sa"1>le size determination is one of the many important issues in the planning stage of a clinical trial. As more successful therapies emerge, clinical trials are increasingly moving from efficacy tests to equivalency tests, in an attempt to reduce side effects or improve quality of life while maintaining clinically equivalent response rates, A SAge program haS been developed to calculate the sample size for equivalency trials with binomial outcomes. The user indicates the type I error rate (a). the type II error rate (13), the r:esponse rate of each of the treatment groups (It 1 and 112) and the allowable difference between the groups (5). Optionally, the user can elect to calculate the power given the sample size andlor select a preset range for a, P , 1t2 and Ii. The output from the program is a table of the estimated samples sizes for the parameters specified. Sample size calculations are illustrated using the examples from clinical trials data.	sample size, sample, clinical trial, interim analysis, significance level, binomial distribution	source	6
7	A SAS Software Program to Calculate Sample Sizes for Comparing Time-to-Failure Among More Than Two Groups	Jennifer Lee, Tamara Odom-Maryon	This software program was developed to calculate estimates of the required total sample size for comparing time-to-failure among two to six treatment groups. The program is based on the formulae developed by Makuch and Simon where time-to-failure is assumed to follow the exponential distribution. The user specifies the following: (1) the number of groups; (2) the type-I error α; (3) the power (1 - β); (4) the maximum ratio of the median/mean failure times; (5) the estimated survival and lost to follow-up rates at a specific time point; and (6) the projected years of patient accrual and follow-up. Under the null hypothesis of an equal number of failures per treatment group, the program provides the total sample size required to calculate the desired differences in group median/mean failure times. An example of an oncology clinical trial is given.	sample, sample size, proc step, %if-%then-%else	source	4
8	An Integrated Clinical Trials System Utilizing Client Server and Graphical User Interface Technology	Martin J. Rosenberg	Motivated by the desire to better manage an ever increasing volume of information and to reduce the length of time required to introduce new drugs around the world, the pharmaceutical industry and regulatory agencies such as the U.S. Food and Drug Administration have sought to facilitate the drug development and approval processes through innovative uses of computer technology. One such effort· that has received great attention is the computer assisted NDA or CANDA and its cousin for biologicals the computer assisted PLA or CAPLA. The first CANDA's were created with much effort, and at great expense, after the paper NDA's had been filed. From these early experiments, it became evident that if CANOA's were to ever become routinely used throughout the industry by both large and small companies, their development would need to be simplified and to begin earlier in the drug development process.	clinical trial, data review, crf, clinical data, audit trail, data validation rules, pharmaceutical industry, key, clinical research, dataset	source	7
9	An Introduction to SAS/FSP Software	Cyndie B. Gareleck, Terry Fain	`SAS/FSP` is a set of procedures used to perform full-screen interactive data entry. data edlting. data retrieval. and data query. This tutorial provides a basic overview of SASIFSP software under version 6.07/6.08 of the ~ System. Topics include: creating a new SAS data set customizing data screens dataently dataquery * browsing SAS data sets and extemal files The focus of this tuto.riaI is on `PROC FSEDIT` although other procedures such as `PROC FSBROWSE` and `PROC FSVIEW` wHi be briefly discussed. Screen control language (SeL) is considered an advanced topic and will not be covered in this tutorial.	proc step, dataset	source	4
10	Building Data-Driven Object-Oriented Applications: Selected Techniques	SAS Institute (a company), Michael Gilman, Donald J. Henderson, Martha F. Johnson, Merry G. Rabb, David S. Septoff, Norman Shusterman, Gregory A. Smith, Robina G. Thornton, Philip H. Van Dusen, C. Jessica Yuan	This tutorial describes a system's architecture in which an application is built using reusable modules or classes of objects. The application can be built, controlled, maintained, and enhanced through changes to the underlying data and parameter files. An application designed in this way allows for more flexibility and lower maintenance than one where the code is written to perform specific functions. Generalized modules or classes of objects, the actions of which are dictated by parameters, lend themselves to the inevitable changes that are made to the system. Many modifications can be accomplished by changing values in parameter files instead of changing the source code. Features of this approach include adding new reports, changing or reordering menu items, and controlling user access and security. A sample application that makes use of `SAS/AF` and `SAS/FSP` software using Screen Control Language (SCL), the `SQL` procedure, and the SAS macro facility is presented and described in detail. The use of a Toolbox library of reusable modules and objects is also illustrated. The data-driven approach combined with the use of reusable tools or objects can be adapted to a wide variety of data-processing tasks of varying complexity. The techniques presented here merit attention as models which may suggest solutions to many system design problems.	sample, proc sql, macro variable, proc step, libname statement, macro loop, macro definition	source	7
11	Comparing Japanese and American GDP-Money Lag Structure Using SAS PROC PDLREG	Reiko Ishii, Ryotaro Tsutsumi, John F. Killeen, Joseph E. Earley	The purpose of this paper is to show the ease and efficiency with which `PROC PDLREG` may be used to estimate distributed lags. Regression models are developed for the United States and Japan with the rate of growth of GDP as the dependent variable and the rate of growth of the money supply as the regressor(s). `PROC PDLREG` is implemented by determining both the order and length of lag of the regressors. The utility of each model is then determined by analysis of the usual criteria, such as regression coefficient t-values and model coefficient of determination. The discussion concludes by comparing the model results for the United States and Japan.	regression analysis, proc step	source	2
12	Cooperative Processing with the SAS System	John McIntyre	One of the four fundamental barriers to SU<:eessful enterprise-wide information delivery is the incompatibility of various hardware and operatingsystemenvironmenls. Customers with enterprise-wide information needs must be able to leverage their investment in technology, both old and new. Many terms are used in the fietd to describe the strategies available to customers including client/server, distributed processing, and downsiZing. This paper will examine models of cooperative processing, review the use of SAS System tools-in particular `SAS/CONNECR` and `SAS/ACCESS` software-to accomplish the integration of resources, and suggest some guidelines for the distribution of application tasks. INTRODucnON Cooperative processing, as a style of computing, commands a considerable amount of attention in MIS installations around the world.	stored procedure, relational database	source	4
13	Corporate and City User Groups	Charles E. Shipp	The author has experience in starting and supporting user groups that include the Northrop SAS Users Group. the Southern California SAS Users Group. and the Northrop Macintosh Users Group. He is a member of the board of directors of the Los Angeles Macintosh Group (LAMG) and started the Special Interest Group connection Corporate (Macintosh) User Groups. The difference between user groups and clubs is explained. and the different leadership styles are discussed. Techniques that build in risk but have potential for spectacular success are analyzed and contrasted with techniques for moderate success but a greater guarantee of continuing success. The paper discusses the benefits of user groups to the corporation. the individual user. the user's department. and to SAS Institute.	proc step, data visualization	source	4
14	Development of an Emissions Modeling System Based on SAS Software	Suzanne J. Strasser, James G. Wilkinson	Emissions estimates of photochemically reactive pollutants are a key input to a photochemical model. For proper photochemical model performance, emissions estimates must be spatially located within the modeling domain, temporally resolved on an hour-by-hour basis for the specific modeling episodes, and chemically speciated. Most emission inventories are compiled on an annual basis for individual counties or planning regions. Radian Corporation has developed the Geocoded Emissions Modeling and Projections System (GEMAP) to process these annual county-wide emission rates into gridded, speciated, hour-by-hour emissions estimates required by current photochemical models. GEMAP is a SAS-based system that calls ARC/INFO programs to perform certain spatial data manipulations, such as development of the grid structure. In addition, FORTRAN routines are used to perform certain numerical functions, such as the computation of motor vehicle emission factors. This system is menu driven and is comprised of over 50,000 lines of SAS code. It uses the following SAS products: Base SAS; `SAS/GRAPH`; `SAS/AF`, including SCL and PMENUS; and `SAS/FSP`. This paper will discuss how each of the above SAS products were used to develop a comprehensive, user friendly emissions modeling system. Methods of integrating the various SAS products will be addressed.	macro variable, proc step	source	6
15	Improving Business Education with SAS: Business Students Perceptions	Miro Costa	The objective of this paper is to investigate how we can improve business students education using SAS, both in PCs and in the mainframe environment. An experiment was conducted at California State University, Chico in 1992 with students enrolled in an introductory course in Management Information Systems. The subjects in this experiment were in management, marketing, finance, accounting, production, MIS and in other nonbusiness majors. The students were exposed to SAS for the duration of that introductory and required 16-week course. A survey was conducted at the end of the course where the students were asked to express their perceptions and expectations about SAS use in the business environment. The results are reported and discussed. The implications of this study may prove to be important for a large number of Business Schools in U.S. and in other countries.	sample, sample size, management reporting, proc step, proc freq	source	9
16	Improving Service Quality With the SAS System	Robert N. Rodriguez, Katherine J. Roggenkamp	Although the American quality revolution of the 19808 began in the manufacturing sector, total quality management (TQM) programs are now common in the service sector. As a result, SAS users in organizations ranging from banks and hospitals to public utilities and government agencies are asking how the SAS System can be used to support data-driven TQM. Quality improvement applications in service settings have requirements distinct from those in manufacturing settings. Service quality is typically determined by customer satisfaction rather than technical specifications, and it involves direct transactions with large numbers of customers and high volumes of transactions. The importance of customer survey data is recognizedl, but statistical and graphical methods for data analysis are relatively underutilized.	customer satisfaction, data science	source	2
17	Is the SAS System a Database Management System?	William D. Clifford	Commercial Database Management Systems (DBMSs) provide applications with fast access to large quantities of data. In addition, many have other capabilities such as data integrity services, data sharing, application-creation tools, and report writing. Version 6 of the SAS System also contains a number of similar features. This paper examines the database features of the Version 6 SAS System and compares them to the services offered by several popular DBMSs. The	flat file, constraint, proc sql, audit trail, data cleaning, where statement, data standard	source	6
18	Keys to the DATA: Using the KEY= Option on the SET Statement	Ben Conner	One of the most efficient and powerful features to come out of SAS release 6.07 is the KEY= option on the `SET` statement. In many cases it rivals the conciseness of SOL for merging data files while retaining all the flexibility of `DATA STEP` programming. This paper discusses the advantages of the KEY = option and the pitfalls to avoid when using it.	set statement, key=	source	3
19	New Directions in Graphics	Jade Walker, Mike Kalt	This paper describes new graphics products, and enhancements to `SAS/GRAPH` software for release 6.08 and 6.09 of the SAS System. New products include `SAS/GIS` (a geographical information system), and `SAS/IMAGE` (which can be used to import and manage graphics images). Enhancements to `SAS/GRAPH` software include the experimental procedures JCHART, which can be used to produce donut charts, and JPRINT, an enhanced version of `PROC GPRINT` with additional line drawing capabilities. Other features under development include a procedure for scientific data visualization, a paint facility, and video processing software.	proc step, proc gchart	source	3
20	New Directions in Technical Support	David C. Brumitt, Sally J. Painter	At SUGI 17, the Technical Support Division announced four new support services available to SASe software installations. These three new features, in addition to the existing OnOne Customer Support Facility, offer a full range of electronic support services. The electronic support services available are: * Electronic Mall Interface to Technical Support	proc print, sas log	source	5
21	Programming for Job Security: Tips & Techniques to Maximize Your Indispensability	Arthur L. Carpenter	A great deal has been said about programming techniques for efficiency and maintainability of SAS'" programs. We are taught to write code that minimizes machine and programmar resources. Unfortunately. easily maintained code requires fewer programmers. and fewer programmers means pink slips and less job security. In these troubled times. programmers naed to be able to maximize their indispensability. Programmers can increase their job security and thereby protect themselves and their families by applying the tips and techniques discussed within this paper. The progremmar will be edvised when to apply the techniques. and whether the use of the techniques should be subtle or gross. Techniques will cover programming style. editing style. statements to use and avoid. and naming conventions. You will learn to blur data steps, make non-assigning assignment statements, and in general write code that not even you will be able to figure out how or why it works.	drop statement, job security, macro definition	source	4
22	Random Sampling Without Replacement	Hong Zhou, Patrick Romano	This paper presents a simple approach of using SAS DATA steps to do random sampling without replacement and provides SAS programs for three applications of ran dom sampling without replacement. These applications are random permutation, random splitting, and random selection.	sample, random sampling, bootstrap, proc sort, randomization, data cleaning	source	2
23	Recruiting, Interviewing, and Hiring a SAS Programming Staff	Judith H. Mopsik	Would you buy a car without a test drive? Would you hire a juggler without an audition? Then why would you hire a SAS programmer without an audition? During the last fourteen years, Atlantic Research Corporation (ARC) recruited hundreds of SAS programmers while building several large centers of SAS expertise. Our goal in the recruiting process is to determine whether or not a candidate has the potential to become a strong member of ARC SAS programming staff. We followed the traditional hiring approach: senior staff spent large amounts of time screening, interviewing, and hiring prospective employees. Six months later, we reviewed the new employee job performance and found that we had not always accurately assessed the new hire potential.	sas programming, proc step	source	7
24	SAS Consultant's User Group - CONSUG	Ben Conner	This paper discusses the recently formed SAS Consultant's User Group. It covers its goals and	sas consultant, service provider	source	5
25	SAS Institute's Training Options for the SAS System for Information Delivery: Overcoming the Final Barrier	Marti Dominick, Deborah Upchurch, Bonnie Allen	The barriers to true enterprise-wide information delivery that once existed are coming down. Release 6.08 of the SAS System is available across multiple hardware platforms and operating systems. New SAS products and enhancements to existing products have strengthened the SAS System, making effective use of the software for your information delivery needs more critical than ever before. Many of you consider the SAS System to be world class software. We at SAS Institute are not content just to provide great software; we also want to provide you with world class training. The SAS System for Information Delivery gives you the tools you need to overcome the various barriers between you and your information. SAS Institute also provides comprehensive training to help you become immediately productive with the Information Delivery System.		source	3
26	SAS Oriented BBS Systems: Online Help 7 Days/Week	Ben Conner	This paper describes electronic Bulletin Board Systems (BBS) and how they can be useful to the SAS user community. It covers character istics of BBS systems in general and what differ ent ones can do. Finally it describes 2 BBS sys tems that cater to SAS users.	group by, delimited file	source	7
27	Successful Simultaneous Facilitation of SAS in the City of Phoenix, Including the Development of Documentation and Design Methods	Michael L. Sperling	A pilot project allows lime for discovery. Discovery of many things ... how to do specific functions, how not to do certain other functions, how to be sure that the successful items continue to be successful, and how to make successes from those items which went awry. This pilot was no different. We .have had several things which did not work the way we thought they would or should. However, overall, the pilot has been a success. One of the most important successes of this pilot was the forming of a partnership between the Risk Management Divivsion of the Finance Department and the Management Information Systems Department.	sas consultant, batch job	source	6
28	Table Lookups using PROC FORMAT, the PUT Function, and a LOOKUP Macro	R. David Barsky	This tutorial describes one simple technique for implementing a lookup table: `PROC FORMAT` to creala the table and the PUT FUNCTION to access it. A short overview of the reasons for and advantages of using lookup tablas, a review of both `PROC FORMAT` and the PUT FUNCTION, alternatives to using a lookup table, and a reacly-tl:Hlse MACRO namad "LOOKUP' for creating a lookup table from an existing SAS data set	lookup, proc step, proc format, put statement, dataset, data cleaning, format statement	source	2
29	Temporary Arrays: An Alternative Approach to Merging Large SAS Datasets	James G. Wilkinson, Suzanne J. Strasser	We have developed a method of merging large SAS" data sets that has resulted in impressive improvements in CPU performance as well as significant reductions in disk storage requirements through the use of temporary arrays. The SASII> `MERGE` statement involves sequentially reading and writing observations in SAS" data sets. This method is extremely slow with data sets containing thousands of records. The new	dataset, cpu time, temporary array	source	4
30	The Efforts of the Process Improvement Committees at SAS Institute Inc.: What Works?, What Doesn't?, Who Cares?	Edie Jeffreys	People at SAS Institute are always actively looking for methods to improve our lives and the lives of our customers. In the last few years, that pursuit prompted the formation of new organizational groups focusing on processes within the Institute. The idea of company-wide process improvement was initiated when Dr. Jim Goodnight, the president of SAS Institute, decided that ideas for software process improvement being presented in an ad-hoc fashion in upper management meetings would be better served in the Software Quality Assurance (SQA) meetings. Attendance at the SQA meetings jumped significantly. For the first time since the corporate divisions were split up into different buildings, ideas for improving the overall software production process were being fluently shared across organizational lines.		source	5
31	The LACKFIT Option in SAS LOGISTIC Procedure: A Problem and a Proposed Solution	Estie Hudes, L. Palermo	There are some useful new options which enhance the `LOGISTIC` procedure for SAS 6.07 (SAS Technical Report P-229, Release 6.07). A very important one, which tests for the goodness-of-fit (GOF) of the model, was suggested by Hosmer and Lemeshow (Applied Logistic Regression, Wiley 1989). However, there are a few problems with the way this was implemented by SAS Institute: The LACKFIT option works well when the model statement is in the event/trials syntax. It does not work properly in conjunction with the new FREQ option for grouped data. It also does not work properly for ungrouped data, when each case is an observation in a SAS data set. The LACKFIT option produces results which are different from the results for the count version of same data set. Finally, the LACKFIT option does not work properly with a WEIGHT statement. We wrote a macro which takes the output from `PROC LOGISTIC` and produces the correct GOF statistics for both ungrouped and grouped data. We also have a version which deals with weighted data. A copy of the macro code is enclosed in the Appendix. A WEIGHT version is available from the authors.	logistic distribution, proc logistic, proc step	source	2
32	The Urge to Merge: A Computational Method for Linking Datasets With No Unique Identifier	Robert M. Bell, Joan Keesey, Toni Richards	This paper describes a procedure used to link California vital statistics records for very low birthweight (VLBW) infants (weight less than 1500 grams) to Medicaid claims data. Because the two main data files share no unique identifier, record linkage required combining evidence across several identifying variables such as name, birth date, and delivery hospital. We used record finkage theory to compute scores that measure the likelihoocl of a true match, I.e., that two records correspond to the same delivery. These scores weight appropriately the various pieces of evidence for or against a match. The scores allow for missing data, data errors, and other factors that may cause a true match to disagree on some variables. The resulting procedure is efficient, accurate, and noniterative. It eliminates the necessity of repeated merge steps and makes the most efficient use of the matching information, thereby minimizing the false positive rate associated with any number of presumed matches.	record linkage, sample, unique key, proc freq, proc step	source	6
33	Training People to use the SAS System under UNIX Operating Systems	Mark Schildhauer	The SAS System running under various forms ot the Unix Operating System may currently represent the ultimate combination of power, affordability, and ease-ot-use for sites seeking to 'right-size' their configurations. Whether migrating up from PC's, or scaling down from mainframe systems, SAS on .Unix is worth a look. as it has the price and ease-of-use advantages oJ the former. ano the performance advantages of the latter. At our academic site. the benefits of running SAS on Unix boxes has led to a rapid migration of end-users away from both mainframe and microcomputing environments. In this paper I discuss some of the issues involved with helping users make the transition from some other operating system to Unix. Ironically. this may involve convincing users ot the advantages to working on Unix systems in general. as well as informing them of the specific performance benefits of running SAS under Unix.	txt, exploratory data analysis	source	9
34	Transferring SAS Datasets Between Your PC or Next and the VAX	James Ssemakula	This paper gives a detailed description of some ways of transporting SAS datasets between a Personal Computer (PC) or NeXTstation (NeXT) and the VAX cluster at a specific site, the University of California, Riverside (UCR). In fact the main body constitutes a handout given to users to help them in the transfer of SAS datasets. Obviously, it is geared toward novices and can easily be modified to suit the needs of other sites.	dataset, proc step, proc cport, proc copy, proc cimport, libname statement	source	10
35	Using SAS Software to Implement a Simple Method for the Analysis of Clustered Binary Data	Tamara Odom-Maryon, Chul W. Ahn	Clustered binary data occur frequently in many fields of applications. For example in the field of cancer, diagnostic radiologic tests result in the identification of multiple lesions for each patient. The dependence between these observations within a patient (referred to as a cluster) makes the use of usual chi-squared statistics invalid. Recently, Rao and Scott (1992) developed a method for comparing independent groups of clustered binary data with group-specific covariates. Their method assumes no specific models for the intracluster correlations. After "pre-processing" the data using the concepts of design effect and effective sample size used in sample surveys, the standard procedures in `SAS/STAT` software for the analysis of independent binary data can be implemented. A program to perform the "pre-processing" of the data as well as an application involving clustered binary data are presented. Further applications to clustered data from a clinical trial in oncology are underway.	sample, design effect, binomial distribution	source	5
36	Using SAS to Deliver Census Information	Dennis X. Tang, Ying Zhou, Terry Yergan	Southern California Association of Governments (SCAG) is a designated Census data center for the southern California region. As such, it is responsible for dispersing Census data and furnishing customized reports requested by individual end users. These customized reports are . either hard copies, maps, or computer files on diskettes or tapes. At SCAG, SAS is used to carry out the Census data center functions on the IBM RISC6000 Unix platform. This paper will examine in detail some of the SAS progi'amming techniques used.	proc step, dataset	source	7
37	Using the SQL Procedure	Kirk Paul Lafler	The `SQL` procedure follows most of the guidelines established by the American National Standards Institute (ANSI). In its current form, `PROC SQL` provides many features available in the `DATA STEP` and the MEANS, PRINT, and SORT procedures. An advantage of using `PROC SQL` is that it can often result in fewer and shorter statements than using existing `DATA STEP` and procedure methods. This hands-on workshop illustrates numerous examples of how `PROC SQL` and its many statements can be used. In particular, participants willieam how to create and modify tables and views, retrieve data using the SELECT statement, and perform efficient queries against SAS System data sets. Brief History of Structured Query Language (SQL) Structured Query Language (SQL) has an interesting history: It is a universal language that was developed to easily access data that is stored in relational databases or tables.	proc sql, group by, order by, proc step, create table, relational database, update statement, where statement, select	source	10
38	Visualization and Analysis of Disease Progression Over Time	Joseph I. Bormel, Linda R. Ferguson	Analyzing raw data can be prohibitively time consuming. A variety of graphical tools have been developed to address this problem. Although graphical analysis provides a simple yet comprehensive overview of a large dataset, often this technique fails to capture the essence of data trends. In addition, the ability to quickly query any component of the data subset frequently remains burdensome.	proc freq, proc step	source	3
39	Visualization of N-Dimensional Databases Using Parallel Coordinates	J. J. Helly, Arthur L. Carpenter	Conventional methods of exploratory data analysis employ tools ranging from two and three dimensional scatterplots and descriptive statistical summaries to multivariate methods employing factoring and model formulation. Much of this kind of effort is directed toward distilling the important parameters and observations from noisy, and sometimes sparse, databases. Parallel coordinates provides a method for examining the relationships between variables in the context of the complete set of observations. Unlike conventional graphical methods, the parallel coordinate method is not restricted to three dimensions or four if one considers animation. Since parallel coordinate representations are based on the methods of projective geometry, fundamental relationships such as positive or negative correlations between variables can be discerned directly from graphs in a manner similar to conventional scatterplots but for more variables simultaneously. This results in a more compact and efficient tool for browsing and familiarizing oneself with a complex database with practical benefits in quality assurance and analysis of databases.	exploratory data analysis, outlier	source	4

Back to Conf go 1994

This site serves as a convenient hosting location and does not work as a publisher. All copyrights belong to their respective owners.

The information on this site is provided for general reference purposes only. Whilst every effort is made to ensure that the site is up to date and accurate, no warrant, nor any responsibility or liability for, the accuracy or completeness of the content or for any loss which may arise from reliance on information contained in this site is provided. Links to other sites are provided for your convenience but no responsibility or liability for the content of those sites or of any external site which links to this site is accepted. References to any specific commercial product, process, service, or corporation are also for information only, and do not constitute endorsement, recommendation or favoring.

SAS is a registered trademark of SAS Institute Inc. SAS, Statistical Analysis System, and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc. in the USA and other countries. ® indicates USA registration.