SCSUG 1991 Papers - Index

24 papers. Source: SCSUG (South-Central SAS Users Group) conference proceedings.

go 1992

#	Title ×	Authors ×	Abstract ×	Keywords ×	Pages
1	A Variable Selection Algorithm for Quadratic Discriminant Analysis	Samuel L. Seaman, Dean M. Young	One difficulty in the application of statistical discriminant analysis obtains when the researcher must select the most important variables to be included in a statistical analysis. Typically, the relevant variables have been identified using some objective criterion from the substantive discipline in which the application is made. However, the resulting set of variables can contain redundancies, making a further reduction in variables desirable. Indeed, there are other reasons that a researcher might wish to perform variable selection for a statistical discrimination algorithm.	feature selection, discriminant analysis, covariance, sample	4
2	All About Strings	Rhena Seidman	Understanding a few details about how SAS processes character data can help tremendously in avoiding unexpected puzzling results.	character variable, data step	5
3	Analyzing Repeated Measures Designs: What You Don't Know CAN Hurt You	Richard W. Cole	The analysis of data having repeated measures for each observation is typically analyzed using an ordinary least squares model known as Repeated Measures ANOV A where, among other things, equal change across time points is assumed for the observations within a group. When this assumption known as additivity is not met, change across time is likely to be occurring in different amounts for different observations within a group. This "differential change" model is, in reality, very common when the effect of some intervention is being examined. For example, persons with low scores initially, tend to benefit more from an intervention than those having already high scores. This paper examines ordinary least squares regression models for fitting and testing data having repeated measures having two time points. Examples applicable to various behavioral sciences are presented and discussed.	repeated measures, null hypothesis, anova, proc reg, response variable, sample, hypothesis testing, main effect, proc step, regression analysis	12
4	Bringing in the New Stuff: Introducing Release 6.06 to a Large-Scale, Diverse, End-User-Oriented IBM Environment	Kathe Richards	Inttoducing Release 6.06 to a Large-Scale, Diverse, End-User Oriented IBM Environment	proc step, proc sql	5
5	Building Code That is Easy to DeBug and Maintain	Jimmy A. DeFoor	This paper discusses the essential elements of easy-to-maintain code and describes how a full-screen editor can be best used to modify code.	dataset, if-then-else, proc format, proc step	17
6	Calculating the Power Function of the Wilcoxon Rank Sum Test in 2xK Tables	Mark Von Tress	An `IML` program is described that calculates the power function of the Wilcoxon Rank Sum test in 2xk tables. The program may be used for sample size requirements in experiments involving 2xk Tables. 1 Introd uction Many clinical trials use ordinal categorical responses as primary efficacy variables. The ordinal response is used to categorize a subjective measurement. Typically the scale is accompanied by a detailed description of each category which helps reduce misclassification error. In parallel group clinical trials, two treatments are given in a blinded fashion, and the investigator, or patient, is asked to rate the response on the ordinal scale. The Wilcoxon rank sum is a well known choice for determining treatment differences in such clinical trials. The power function for this test statistic must be used to determine the sample sizes for these trials. The problem is slightly more complicated than designing a study based on the t-test	power analysis, sample, beta distribution, sample size, clinical trial, proc freq, prior probability	7
7	Confidence Intervals for Bionomial Distributions Using SAS Graphics and PROBBNML	Alan Davis	An iterative program is shown which determines conTidence intervals Tor the estimate OT a binomial probability parameter PI. Comparisons are made with the normal approximation oT the conTidence interval. For n=20, alpha=.05, graphs show the characteristics oT these conTidence intervals. A second program is provided Tor determining the expected degree or conridence ror speciric values or PI. These programs can be run in a SAS `DATA STEP` and can be easily edited to any n or alpha	binomial distribution, confidence interval, sample	7
8	Considerations for a SAS 6.06 AF/FSP and DB2 Application	Kent Andrew Spencer	The new version of SAS gives developers powerful links to DB2. NOw, manipulating information in third party data bases is nearly as easy as manipulating information in SAS data bases. SAS `PROC`'s for producing graphs and statistics can now access dynamic DB2 tables. Unfortunately, we are not dealing with true SAS data bases, and certain limitations exist. This paper discusses the issues address in developing Halliburton Company's Employees' Trust contribution Reporting System	likelihood function, key	3
9	Empowering the User to Submit DB2 Reports	Gerald Muffoletto		sas output, ledger, proc freq, proc step	39
10	End-Users Training at a New SAS Site	Twila Faye Turpen		dataset, sas programmer	10
11	Evaluating Efficiency Techniques Used in the Processing of Large Data Files	Alfred W. Ey	The processing of large data files can often strain system resources. There are several techniques suggested by SAS Institute which can help to improve processing efficiency and resource utilization. These techniques often call for a tradeoff between time and space and may be efficient only under certain conditions. This paper evaluates some of the methods used to optimize CPU time, I/O, main memory, and disk storage. Each method of optimization is tested under differing conditions and analyzed as to th'e amount of efficiency gained versus the resources needed	cpu time, if-then-else, proc freq, where statement, proc step, frequency analysis, sample, proc univariate	17
12	Evaluating, Programming, and Managing Successful SAS Projects	Tom Mannigel		input statement, proc greplay	13
13	Multiple Engine Architecture and the Base Engine for the SAS System	William D. Clifford	The SAS System supports access to data in a wide variety of formats with its Multiple Engine Architecture (MEA). The. base engine for SAS software defines the format of data for all applications and is responsible for creating and accessing SAS data sets. This paper examines important features provided by the original base engine in Release 6.06 and describes enhancements made for Release 6.07. Included are feature and performance comparisons among Releases 5.18, 6.06, and 6.07, of the SAS System		8
14	Nearest Neighbor Hot-Deck Imputation for Missing Value With SAS/IML	Thomas W. Sager, James P. Gise, M. W. Hemphill	Dealing with missing values continues to challenge statisticians. In this paper we examine the application of one modern missing value technique, nearest neighbor hOI-deck imputation (NNHDI), to one large data set in which about 15 percent of the data are missing or incom plete. Extensive computer processing is required, for which SAS `IML` (Interactive Matrix Language) provides a compact implementation. Each observation in the data set is a vector of related measures, of which one or more components may be missing. NNHDI involves imputing, or filling in, missing components in a target observation from a complete donor ob servation whose components closely match the nonmissing components of the target. `IML` code, provided in the text, is used to compute similarity indices between observations, search for close matches, and impute values from the donor observation to the target observation	missing value, imputation	9
15	Operations Enterprise Model	Richard W. La Valley, Neil H. Howard		quality of service, account	10
16	PROC AUTOREG to Forecast With a Time-Series/Cross-Section Autocorrelation Model	Neil S. Fleming, Willy Waks		proc autoreg, response variable, proc reg, proc step, statistical software, regression analysis	16
17	PROC CORRESP: Different Perspective for Nominal Analysis	Richard W. Cole		proc corresp, chi-square test, proc step, proc freq, likelihood function, proc princomp	8
18	Taming the Beast: A Software Tool for Managing Field Verbosity Using SAS Macros and SCL	Betty Smith		macro facility, macro variable	26
19	The ALLOTT Model: A PC-Based Approach to Demand Distribution for Siting and Planning	Elise M. Bright, Parviz Nazem		demand forecasting, sas programming	12
20	The General Structural Relation Through SAS/AF Windows: An IML Macro You Can't See	Robert J. Zambarano		normal distribution, standard deviation, regression analysis	19
21	The SAS Macro Language: An Overview	Thomas J. Winn Jr.		macro variable, macro facility, macro definition, automatic macro variable, sas programming, &variable, proc step	6
22	Tips, Tricks and Traps, Part 1	Neil H. Howard	Sivilll money i.uppermo.l iJ1111e miJ1dlo! DP m8l138ers. Inelile SAS Syllem i. c:enail11y a "programmer efficienl" 1001: minimal SAS code is requiree 10 perlorm IradiUonaJIy code inlell.ive lists. GralificaliOll IIId IIId resulll are immediale. A familiar qUOleou .. e.lllballbe SASSysiem iseasylOuse.lIld jusl al easy 10 abuse. Because Ibe SAS SYllem i. becomilla illcl"easiJ1aly popular as a developmelll 1001. it is essential 10 evatuale lecb.D.iques lor ill efIeClive use. Tbis IUloria! will deliJ1e efliciellCY, .ullesllaClorslo be consideree ill measurilla efficiency, and provide comparisolll ct lecb.D.iqUeslbal improve etticiency. A primary area wbere Ibe SAS Syslem can be used more efleClively is 110. II sbould be nOled Ibat over 90 s of Ibe `PROC`:essiJ1llime in allY given program Ulvolves reading and/or writillg dala (reference I). Tbi. IUlorial will illuslrale Ilow 10 avoid Ul1l1ecessary 110, milllmiZe pass ** 0[ Ibe data. and conlrol read-wrile loop *	proc step, conditional do loop, proc format, if-then-else, proc append, proc means	6
23	Transferring SAS Datasets Between the PC and the Mainframe	James Ssemakula	This paper gives a detailed description of the implementation of the SAS Micro-to-Host link at a specific site. In fact, the main body of the paper constitutes a handout given to users to help them in the transfer of SAS datasets. Obviously, it is geared toward users who have only a limited knowledge of computers. It should be easy to modify this handout to suit the needs of other sites. Although the primary focus is on DOS and MVS, much of it remains true for other environments and for `SAS/CONNECT`. The handout describes a convenient way of transporting SAS datasets between a Personal Computer (PC) and a host mainframe computer at Texas A&M University (TAMU). The versions of SAS it covers are 5.18 on the IBM 3090-300 mainframe (MVSlESA) and 6.04 on the PC (DOS 3.0 or higher). Read the entire document before commencing file transfer. To communicate with the mainframe, your PC and/or communications package should conform to the TAMU communication protocol parameters	dataset, proc step, proc download, proc upload, sas library	9
24	Using the SAS System Under AIX Effectively	Maria Moore	This paper discusses issues involved in tuning the per formance of the SAS System under the AIX operating system. In order to optimize performance, both the allocation of hardware/software resources and the set ting of SAS system options must be considered. In cluded are discussions on SAS system options, hard ware configurations, 1/0 constraints, disk usage, mem ory size and usage, and network traffic	options statement, cpu time	8

Back to Conf go 1992

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.