HOME QUICK SEARCH AUTHORS KEYWORDS HELP BUGS

SUGI 1981 Papers - Index

109 papers. Source: SAS Users Group International conference proceedings.

Back to Conf
go 1982
go 1980
# Title
×		 Authors
×		 Abstract
×		 Keywords
×		 Extras Pages
1 A Comparison of Graphics Packages Used with the Geoecology Data Base C. J. Emerson,
Martha K. Nungesser		 proc gmap, data science source
2 A Comparison of Several Approaches for Calculating LC50s for Binary Responses from Toxicological Experiments Charles M. Bundrick,
Jerry L. Oglesby		 sample, binomial distribution, confidence interval, moving average, goodness of fit, regression analysis, likelihood function source 8
3 A Fisheries Data Base Utilizing SAS Leanna Pristash,
George Roth,
Dan Watson		 In order to completely describe the biological activity in a body of water~ it is often necessary to perform a wide range of sampling efforts. The diverse nature of the data result ing from activities such as trawl sampling, benthic sampling and mark/recapture studies presents serious problems in designing a data base which is both flexible and easily used. The Ecological Services Department of Texas Instruments Incorporated has designed and implemented a data system for fisheries data, based upon SAS data sets and SAS software. By making use of the ability of SAS to emulate a hierarchical file structure, it is possible to split the data into "levels" of data resolution, which can easily be recombined in a number of different ways to fulfill report and analytical requirements. The system naturally leads to the creation of a "data dictionary II , which serves as a common base of communication between biologist and programmer. Significant sample, data wrangling, dataset, data dictionary, phase i source 4
4 A Macro for Two-Population Nonparametric Univariate Discriminant Analysis William A. Blattner,
Donald J. Henderson		 sample, discriminant analysis, cumulative frequency, probability distribution source 6
5 A Mid-Range Sensitivity Planning Model for Dyes and Chemicals Patricia A. Hermes,
James M. Watts		 dataset, missing value source 6
6 A SAS Job Stream Generator M. L. Wray,
M.S. Hansard		 data dictionary, proc step, input statement, proc freq source 4
7 A SAS Macro for Calculating the Quadratic Discriminant Function Charles S. Davis The prob1em of assessing the goodness-of fit of a random sample to some hypothesized distribution is cOQsidered for three i~ortant ~~actie.l situations: when the hypothesized distribution F(X) is continuous and completely specified and when F(x) is the normal or the exponential distribution, with parameters to be estimated from the data-. Sever.al of the more common typee of goodness-of-fit tests are briefly described and compa~ed; the strengths and weaknesses of methods currently available to the BAS usar are a1ao diacusaerl. MOdified tests baaed on t~ empirical distribution function (EDF), namely the Kolmogorov-Smfrnov statistic D, the Kuiper gtatiatic V, the Cr~er~on Mises statisti~ W2* the Watson statistic U2 and th~ Anderson-Darling .tatistic A2, are recommended. kDF.statistics are ea~ily calculated, require only one line of 8ignifi~~nee points for each situatian and are competitive in terms of power sample, by group processing, probability distribution, normality test source 7
8 A SAS Macro for EDF Goodhess-of-fit Tests Charles S. Davis An important assumption underlying the use of the linear discriminant function is homogeneity of the within-group covariance matrices. In practice. this equal covarian<;e assumption is often not satisfied. The facility for quadratic discrimination, 'Which ~es not . require the assumption of equal covar~ances, 48 a major advantage of the SAS DISCRIM procedure. When the pooled covariance matrix is used in the discriminant model, the linear discriminant function is always printed. Unfortunately, when the within-group covariance matrices are used, the quadratic discriminant function (QDF) is not similarly available. A macro, utilizing PROC MATRIX and the output data set created by PRQC DISCRIM, for computing and printing the QDF is described covariance, proc step, sample, discriminant analysis, proc discrim, sample size source 6
9 A SAS Macro for Generating Patient/Subject Random Allocation Schedules James A. Bolognese A SAS macro to generate random allocation sche dules of sequential patient/subject numbers for blocked or unblocked studies in clinical/precli nical trials with any number of crossed factors is presented. The rank transformation is used with SAS' UNIFORM function to produce a nicely formatted allocation schedule. The theoretic justification of the II ran domness" of the rank transformation is given and verified by a Monte Carlo example. In addition, examples of the use of the macro are presented sample, monte carlo, macro definition source 3
10 A SAS Response Surface Design and Analysis Package Frank E. Harrell Jr. proc genmod, proc step source
11 A SAS Routine for Modifying the Output of PROC CHART to Meet Special Report Requirements William M. Taylor class variable, proc step, proc freq, clinical trial source 4
12 A Triad of SAS Macros to Capture the Output from PROC GLM Roger S. Cohen proc genmod, proc glm, proc step, regression analysis source 6
13 Alternative Geographical Display Strategies for Bivariate Relationships Michael P. Farrell,
Rodney H. Strand,
Joyce C. Waterhouse		 missing value, confidence interval source 5
14 An Analysis of Methods Used in ANOVA with Missing Cells Ronald R. Hocking The controversy about the computer analysis of unbalanced ANOVA, althougn not clearly Te solved. 1s better understood than it was ten years ago. Although this problem technically includes, as a special case, the possibility of missing cells, the latter problem is not so clearly resolved. The purpose of this paper is to review some of the more common prQcedures u$ed with unbal anced data and to attempt to clarify the ~aaing cell methods currently used. Given an UDde~$tanding of the implications of existing methods, it ia hoped that futu~e comr puter packages will provide more flexibility and eneouragement to the user to preBer~be the an $.lysis most appropriate for the problem at hand. 1 anova, main effect, proc genmod source 6
15 An Enhancement to NPAR1WAY: Multiple Comparisons Daniel M. Chilko,
Gerald R. Hobbs		 kruskal-wallis test, sample source 2
16 An Interactive SAS Program Writer Applied to Managing a SAS Data Base John B. Carney Jr.,
Guy C. Grenier		 put statement, proc print source 5
17 Analysis of a Two-Period Crossover Design Using Macro XSOVER Irving K. Hwang anova, residual, treatment effect source
18 Asymmetry and Regression Philip Harris Monchar correlation, regression analysis, predictor variable source 5
19 Automated Management of the Regulatory Aspects of the Development and Registration of Pharmaceutical Products F. A. Adornato,
D. K. Jackson,
R. A. Michalak,
C. A. Simon,
C. S. Snoddy		 quality assurance, crf source 3
20 AUTOSAS: A System for Automatically Building and Executing SAS Programs Peter Errington source 6
21 Bibliographic Information Storage and Retrieval System (BISARS) Manjeet S. Chhinnan,
Leonard C. Moon		 Service;, CAIN issued by National Agricultural Library; COMPENOEX issued by Engineering Index. Federal Agencies like DOE and NASA has several information storage and retrieval systems serv ing the same basic purpose as BISARS. These data bases are very comprehenSive and exhaus tive, and require programmer-analysts to main tain, upgrade and use such systems. BISARS is based on simple SAS procedures and can be used with no or very little background in computer programming. The SAS system is currently availab1. on the University of Georgia IBM computer in Athens. This system is accessible both through batch and interactive mode. 317 proc step, proc sort source 10
22 Color Graphics Hardware: Review Morris L.Samit bar chart, line chart source 5
23 Community Information Indices Michael P. Farrell,
C. Pennington,
H. Polovino,
Rodney H. Strand		 sample, proc freq, proc princomp, proc step, correlation source
24 Control Charts with SAS and TSO Dale F. Kraemer,
Judy A. Stober		 sample, standard deviation, quality control, statistic, macro definition source 6
25 CPMS: A Project Management System Implemented in SAS79 James W. Bash,
Linda J. Maher,
David A. Merrill		 proc step, data step source 4
26 Data Management Problems with a Large Medical Survey Henry A. Anderson,
Wm. Ford Calhoun,
Douglas Dorph,
James Holt		 data governance, sample, quality control, audit trail, dataset, repeated measures source
27 Diagnostic Techniques in Multiple Linear Regression Using PROC MATRIX Ronald R. Hocking,
O. J. Pendleton		 regression analysis, proc step, multiple regression, proc reg, influential observation source 7
28 Discriminant Analysis Tom R.Bohannon discriminant analysis, covariance, proc step, proc discrim, sample source
29 Enhancements in Logistic and Proportional Hazard Regression Procedures Frank E. Harrell Jr.,
Kerry L. Lee,
Ray A. McKinnis		 N.w .n.l~tic.l c.pabilit1.. have b n addQd tQ BAS prae.du~.. LDQIST (fo~ loti.tic binarv regr ion) .nd PHOLM (f~? Cox p~GPo?tiDn.l hAzard g.n.~.l lin"T model', The.e include 'he obilitg to t.st tho glabol hgpothesis Ho:8O ~lthout 1te~ati4nc USing the efficient BCQT. statistic. and the .bi1itV to add and delete variables in tft. model in the o,..d,1'" spectfied blj t:he- u ... ,... Thi. l~tt.l' SEGUENTIAL option ~ .. dHht tho un 0' incamplata p~incip.l component. lagi.tic and PTopo~tiDn.l hazard r.g" * ion. FOT tihe PHGLM pTo(:.duT'et blockh'g ha. be.n iIIdded. Thi. allow'S the "' 1' td abSDrb -the .f-fl-.ct. a' ".ctOl'. that a.,.e not modeled. A n.~ ind.x of model ftt h been inco~por.t.d into both pToceduTas. The~. n8~ f ** tu~.w ~ill .ppea~ in ... n.~ rella~. Qf the prQcedur.~ likelihood function, proc step source 3
30 Estimating the Parameters of the Michaelis-Menten Equation Using SAS Thomas P. Capizzi,
Leonard Oppenheimer		 The Michaelis-Menten equation, a rectangular hyperbola with two ~arameters ~o b~ specified, . occurs prominently 1n enzyme k,netlcs." Often, ~n consulting situations we need to (i) obtain estl mates standard errors, and confidence intervals for the unknown parameters, which define this nonlinear model for one set of experimental con ditions and (ii) compare the parameter estimates under different experimental conditions~ Parame ter estimates for this nonlinear model can be obtained using least square normal theory~ the regular jackknife, the linear jackknife or the weighted ]jnear jackknife. In addition, the jackknife procedures allow one to evaluate differences in the parameter estimates across several experiments. The purpose of this paper is to demonstrate the ease in which these techniques are imple mented using several SAS macros proc calis, estimate, proc step, confidence interval, proc nlin, macro definition, regression analysis, sample source 6
31 Estimation of Fixed Effects in Mixed Models with Unbalanced Data Walter R. Harvey fixed effects, proc calis, proc mixed, covariance, least squares mean, degrees of freedom, predictor variable, proc freq, regression analysis source 7
32 Experience of Applying SAS in Engineering Analysis and Design Rhyn H. Kim proc step, proc genmod, regression analysis, data science, proc glm source
33 Geographic Area Profiling Using SAS John G. Blodgett dataset, proc step, sample source 5
34 Growth Curve Analysis Using PROC TABLES George F. Fraction,
S. Stanley Young		 proc step, main effect, proc factor, anova, two-way anova source
35 How to Use SAS for ANOVA of Mixture Problems Mark L. Tiede Probably the most common use for statistics in the research environment is to fit 1inear models to data obtained experimentally in the lab. The null hypothesis, degrees of freedom, proc genmod, predictor variable, anova, regression analysis source 7
36 Implementing a Computer Usage MIS Richard J. Nicola GTE has made use of various computer usage reporting tools and systems to meet the information needs of managers and technicians at its sixteen MVS data centers and at the corporate level. In order to consol idate this reporting, GTE has expanded the concept of a SAS performance data base into an integrated multiple level Management Information System. The MIS addresses all aspects of tuning, planning and chargeback associated with computer usage and provides information to managers and technicians at all levels capacity planning, data wrangling, management reporting, mis source 6
37 Implementing a SAS Census Data Collection System Alvin M. Best III,
C. Royce Claytor		 data wrangling, input statement source
38 Influential Data Diagnostics and Model Building Techniques Applied to an Air Pollution and Mortality Analysis Diane I. Gibbons,
Gary C. McDonald		 regression analysis, predictor variable source
39 Institutional Research and SAS Stephen Ahrens,
L. Robert Kuhn		 data wrangling, length statement source
40 MACRO DATABASE, PROC JOIN, and Friends: Several New SAS Macros, Procedures, and Functions Arnold W. Bragg Jr. MACRO DATABASE 1s a SAS program for summar izing entire SAS databases. Unlike FROC CONTENTS which summarizes by dataset, DATABASE combines variable, device, OS attributes, and SAS sttr!· butes from each SAS dataset in the database into a single report. PROC JOIN is a SAS procedure which 'joins' two datasets in the relational sense. PROC JOIN is similar to MERGE except that one observation is produced for each possible combination of the BY variables. Eight SAS date functions and nine SAS functions which generate pseudo-random variates are also discussed, with examples. MACEO DATAllASE SAS MACRO DATABASE is a SAS program for summarizing entire SAS databases. DATABASE gives SAS users a rudimentary data dictionary/directory facility. Unlike PROC CONTENTS which summarizes by dataset, MACRO DATABASE combines attributes from all specified SAS datasets or SAS databases into a single output dataset from which reports can be generated dataset, proc step source 6
41 Making Sense of Types I, II, III, and IV Sums of Squares in SAS GLM Douglas C. Bolgiano proc genmod, degrees of freedom source 4
42 Management Information Systems Using SAS as an Aid Stephen B. Sloan proc step, proc print, missing value source
43 Management of Quality Assurance Controls: Decision Support Systems Michael P. Farrell,
Rodney H. Strand		 Environmental research involved in the assessment of new coal conversion technologies must evaluate many biological, physical, and chemical variables to determine environmental acceptability. Inherent in these studies is the use of quality assurance controls to evaluate data base entries, analytical procedures, and program productivity. However, quality assurance controls implemented at the project level are seldom known in detail and lor evaluated when making resource management decisions at the program level. This paper presents a computer based decision support system that assists in management level integration of complex environmental research. The DSS is modeled after an open-ended construct applied to data management systems developed for environmental research programs t i.e., minimal planning effort and a flexible system that can accommodate changing research emphasis quality assurance, sample, data governance source 5
44 Missing Cells Data Analysis under a Constrained "U "Model Dorinda Love Casanova Analysis of data involving missing cells Is of major concsrn to many statisticians. This paper proposes ,an alternate method of dealing wit:h this pToblem. whQreby the null model is applied under a constraint condition, to data with missing cells. "Instead of the traditional generalized inverse J a_partitioned inverse is utl1i~ed. The end ·product is a cell means solution vector which PrQvides an estimate for 411 cells including the missing Ones ~hen a constraint condition is imposed upon the model sample, proc genmod, proc glm source 4
45 Missing Data in Multivariate Linear Models: A Comparison of Several Estimation Techniques James D. Hosking sample, correlation, missing value, proc step, response variable, likelihood function, sample size source
46 Multiple Comparisons for Fixed Effects in Mixed Models Gary Regula,
Judah Rosenblatt		 regression analysis, confidence interval source 5
47 Nonstandard Editing Techniques for Clinical Data Richard A. Enz,
John Sanocki,
Doron Steger		 scatter plot, data dictionary source
48 On the Role of Redundant Information in Error Detection Ronald W. Helms data governance, data wrangling, binomial distribution, outlier source 5
49 On-line Estimation in Mixed Linear Models William M.Sallas proc mixed, likelihood function, fixed effects, restricted maximum likelihood source
50 Parameter Estimation, Hypothesis Testing and Coding Chip Kaliher anova, estimate, regression analysis, binary variable, proc glm source
51 Path Analysis: A Multivariate Technique Margaret A. Chmielewski correlation, regression analysis, partial correlation source
52 Path Analytic Models in Human Quantitative Genetics C. Robert Cloninger,
J. Philip Miller,
John Rice		 correlation, proc step, sample, regression analysis, likelihood function source 6
53 Plotting Confidence Bands Using PROC GLM and PROC REG Curtis Barton This paper presents two programs that allow the user to compute and plot predicted values and confidence bands for a dependent variable in a linear model using PROC GLM or PROC REG. These estimates are computed as a function of one of the independent variables in the model statement, conditional on user specified fixed valUeS for the other independent variables in the model. The predictor variable may be discrete or continuous. The user may specify predicti9n at only particular values of the predictor, or interpolation over the range of t predictor proc step, proc genmod, proc glm, confidence interval, proc reg, response variable, predictor variable source 6
54 Plotting Home Ranges for Wildlife Data Lawrence W. Grimes,
Hoke S. Hill,
Herbert C.Krasner		 A SAS program is given which determines from a set of points representing animal sight ings or locations, the boundary points of the animal's horne rangel computes the area of this home range, and displays the home range plot. The determination of the boundary points, points which form the convex hull of the observations, and the computation of the enclosed area are accomplished within a macro (RANGE) using PROC MATRIX. Graphical representation of the obser vations and home range are produced using PROC GPLOT. Illustrative examples and program listings are presented proc step, dataset source 4
55 Probability Plotting in SAS Daniel M. Chilko,
E. James Harner,
Gerald R. Hobbs		 sample, normal distribution, probability distribution, proc step source 6
56 PROC BACKUP: Creating Raw Data Files from SAS Data Sets David H. Christiansen,
James D. Hosking		 out=, dataset, missing value, proc step source 6
57 PROC EDITOR: A New Design for the 80s Emilio A. Icaza proc step, label statement source 4
58 PROC RANDOMTEST: A SAS Procedure Performing Randomization Tests Robert M. Hamer proc step, null hypothesis, permutation test, randomization source 6
59 Processing 1980 Census Data: SAS versus CENSPAC Frank C. Dilorio,
Kenneth A. Hardy,
Judith M. Poole		 data science, data governance source
60 Processing TESDATA Measurement Files with SAS J. William Mullen The availability of non-sensor based measurement capability 1 as p.rovided by the MS-I08 Channel Event Monitor and the MS-I09 Communications Monitor, has added significantly to the volume of measurement data available to the analyst. Merging of this measurement data with measurement data from the host system, such as RMF data, has become a very desirable capability. This paper describes techniques for the transfer of MS Perfromance Data Files (PDFs) to the host system utilizing the Statistical Analysis System (SAS) for data transfer, analysis and reporting dataset, data science source 7
61 Production of an Annual Operational Statistics Report Using SAS Wayne L. Cornelius macro definition, proc step source 5
62 Quirks in Linear Model Calculations for Unbalanced Data Shayle R. Searle Completion over the past two years of Anno tate~ Computer OUtput for a number of analysis of variance routines has ~evealed situations whe~e linear mcde~ calculations for unbalanced data are sometimes a littie surprising or, at best, some what difficult to understand.. Sucb situations are illustrated with (i) faults in ion al<!orithm for repar~t~rizing with E-restrictions, (Ii) surna of squares rOT E-restricted models, (iii) least squares means and (iv) estimating variance components numeric functions, restricted maximum likelihood source 8
63 Reading Data from an IDMS Data Base Using SAS Richard Nelson,
John Trice		 proc step, dataset, sample source 7
64 Relative Risk Estimation from Epidemiologic Studies of Matched Sets Elizabeth J. Martin,
Judith H. Mopsik,
Linda Williams Pickle		 odds ratio, sample, sample size, correlation source 5
65 Reprint File Management Using SAS Charles S. Davis proc step, correlation, sample, covariance source
66 RIMAS, a Research Information Management and Analysis System Using SAS, TSO and CLISTs Ronald G. Haupert format statement, proc step, audit trail source
67 Robust Measures of Scale in Outlier Detection Paul W. Fingerman sample, sample size, standard deviation, outlier, interquartile range source 6
68 Robust Modeling of Data Using PROC MATRIX Anthony A. Billings,
E. James Harner,
Ronald D.King		 regression analysis, covariance, outlier, bootstrap, sample source
69 Rules We Followed and Wish We Had Followed in Managing Datasets, Programs, and Printouts David H. Christiansen,
Keith E. Muller,
J.C. Smith		 data science, dataset source
70 SAS Applications for Personnel Data Analysis Frank Marangell data governance, exponential distribution source 5
71 SAS as a Pedagogical Tool in Corporate Management Training Nelson M. Fraiman,
Murray Mohl		 regression analysis, confidence interval source 4
72 SAS as a System Programmer's Tool James D. Oleksiw proc step, sample source
73 SAS as an Interactive Information Management System Larry Hoover Anyone ~ho has been around data process~ng during the past decade has witnessed the tremendous explosion in the demand for informa tion. This demand, when played against data processing-s ability to respond, have often become a source of contention between users and the data processing personnel. The scenario that follows is usually one of varying levels of frustration as data processing f already overloaded, tries to accommodate this latest demand--albeit a major new system, a revision of an existing system, or simply a one-time ad hoc report--often at the expense of other development work .. Nowhere -is this dilemma more acute than in the case where the demaHd is for on-line/real-time information. Resolving this type of difficulty is not always simple, but the mess can often be mitigated by selecting the most appropriate software to do the job. In a growing number of these instances, SAS has proven to be an excellent tool mis, data governance source 4
74 SAS Outlier Macros: A Banking Problem James A. Calloway,
William L. Seaver		 outlier, regression analysis, proc calis, proc reg, sample, proc step source 5
75 SAS under TSO; Implementation and Applications Gail Jorgensen,
Franklin S. Young		 proc step, proc format source 8
76 SAS/GRAPH Made Easy: An Interactive Approach Sumi Kobayashi,
Robert F. Kramer		 bar chart, confidence interval source 7
77 SASGEN: A Program to Write SAS Code for Simple Data Analysis Morse F. Kalt label statement, proc step source
78 SASTRAK: A Capacity Planning/ Resource Management System Malcolm E McClain capacity planning, regression analysis source 3
79 Similarity Measures: SAS Macros for Calculating Rank Correlations and Jackknife Estimates of NESS Indices Pauline L. McCreight,
John M. Rinehart		 sample, correlation, proc freq, spearman correlation, macro definition source
80 Simulation of Population Genetics Models with SAS Edward L. Spitznagel One of the most important problems of population genetics is the understanding of multi-locus natural selection in the presence of all the realistic conditione that make it so complicated: truncation selection, linkage, and small numbers of major loci superimposed on a background of multi-locus minor effects. SAS has proved to be an excellent vehicle for exploratory simulations, with successive work data sets representing the generations of an evolving population. The saving of programmer time is substantial t and the SAS program forms an excellent outline for the writing of production programs in Fortran proc freq, proc step source 2
81 Solutions for Two Data Management Problems: Data Definitions and Updating Leonard Blavatnik,
Wm. Ford Calhoun,
Douglas Dorph		 The Research-Oomputing Unit supports com puting for 7 Ph.D. statisticians. Each consults on several projects in basic or c~inical scien ces. The projects vary greaUy in complexity and sh:e. Neverthele.ss, our most burdenBOIllE! data m;magement tasks for most projects are ini tial dafinition of the data (resolving input format specifications, variable labels r and value labels) and updating. If not carefully managed, these tasks oonsume an inordinate amo\mt of SAS programmers I time and cause sched ule slippaqe. We have implemented maJlaqement teclmiques and supporting programs that reduce the burden of data definition and updating. .Our data defi nition system employs a l?Ll program that con verts a keyed data codebook into a SAS program. All steps from codebook to SJ\5 data set, data listing. and summary statistics are handled by a data entry operator--in effect replacing a &AS programmer who would otherwise have to per form those tasks data governance, proc step, sas programmer source 4
82 Some Real-Data Examples for Teaching Analysis for Experimental Designs Kaye H. Fendt,
Ronald W. Helms		 regression analysis, missing value source 7
83 STATAN, a Conversational Interface to Facilitate SAS Usage or It Pays To Get Your SAS in Gear Anne M.Parkhurst variable list, factor analysis, proc step, anova, dataset, proc genmod, data science, main effect, regression analysis, correlation, covariance source
84 System 2000 Statistical Analysis System '81 Kirk Paul Lafler data science, proc step source 7
85 Teaching SAS to Experienced Applications Systems Analysts Chapman P. Gleason SAS can be used effect1ve:ly by experienced progrslIIIlers to 80lve a variety o-f pr.()gr:amming applications. Rowaver, the approach to teaching these individuals is different than the Usual l'SAS Basics II course. This paper discusses the design philosophy, topic outline, and teaching materials used in an in-house SAS course for professional data processors. The applieations of SAS to data pro(!(l;ssing problems in a large government survey or8aalza~1on are discussed. Suggestions on how to organize a topic outline for a "SAS Basics" course for applicatiOTlB systems analysts in other organizations ,will also be discussed sample, proc step, data wrangling source 3
86 Teaching SAS to the Entry-Level Programmer Debbie D. Wilson This paper will discuss the SAS training techniques used -in the training program at Kentucky's Depart ment for Human Resources fOT all entry level programmers. These techniques include, A rso CLlST as a tutorial, an in~house written primer, and 3 sample SAS jobs as examples. Since SAS is such a powerful and useful tool for many applications, the Department for Human Resources teaches SAS to pro grammers, regardless of the responsi bilities they are to assume. These programmers mayor may not use SAS after training, but they are made aware of its applicability, simplicity, and quick result's. The eLlST tutorial forms the core of our SAS training with the primer and sample jobs serving as resource aids. Trainees are able, with a minimum' of personal assistance, upon completion of the tutorial to write SAS programs to fill most simple requ~sts we receive. OUr primary use of SAS is fo~ one-time reports on a wide range of file layouts sample, by group processing source 3
87 The Cell Means Model: Evaluating GLM Type III and IV Sums of Squares for Linear Models with Missing Cells Richard L. Miller proc genmod, sample, proc glm, parameter, fixed effects, proc step source 8
88 The Construction of Literature Indexes Using SAS Paul J. Von Doehren Index preparation is often useful when a moderate-ta-large number of literature sources are available for a particular application. Simple alphabetical listings, perhaps with subgrouping by topic, are usually adequate when the number of references is small (e.g., up to 50 sources) and the number of topics is restricted. An improved form of indexing becomes attractive as the numbers of references, topics and potential users increase. Several indexing algorithms already exist. However, the acquisition and maintenance of an existing indexing software package for use with moderate size source lists may not provide sufficient flexibility, simplicity and convenience compared to an algorithm implemented in a familiar high-level language (e.g., SAS). A collection of SAS statements that constructs a listinq of references according to subject is presented. The subject listing is produced using a combination of keyword and keyword-in context approaches to index construction dataset, variable source 7
89 The Design of a SAS Test Score Management System for Richmond Public Schools Ann S. Allen,
Alvin M. Best III		 phase ii, phase iv source 5
90 The Effect of Sample Size on the True Alpha Level with FUNCAT: A Simulation Study John L. Stedl FUNCAT analyzes categorical data using thH Grizzle-StanneJ:~Koch methodology. This method ology ~loits teat statistics which have asymp totic ~ distributions in arder to carry out tesLs of signifieanee. A simulation program was writt.en to investigate the effect of sample size On the true (] levels for var.ious 2 x 2 x k tables when the null hypotheses was true. This simula tion program exaruined two response functions~ log linear and linear. Many true ~ levels were much greater or much less than the nominal levels of 0.01, 0.05 and 0.10, unless the sample sizes were very large. The results indicate that this methodology may not be appropriate for a 100 patient clinical trial with 2 treatments, 2 responses and 10 investigators. Cochran-Mantel Haenszel analysis or ordinary X2 analysis, how ever", of the sam-e simulated data achieved true a levels close to the nominal levels of 0.01, 0.05 and 0.10 sample, sample size, null hypothesis, clinical trial, proc step source 4
91 The Implementation of SAS via SUPERWYLBUR at Air Products and Chemicals, Inc . Thomas J. Bzik,
T. J. Cenna		 dataset, regression analysis, anova, sample source 5
92 The Role of SAS in Influencing Major Transportation Decisions for a Large Private Employer Robert E. Stammer Jr. horizontal bar chart, bar chart source
93 The Statistical Consultant's Dilemma: Providing Help at the Level of Need or the Level of Knowledge John W. Philpot sample, discriminant analysis source 4
94 The Use of Least Squares Splines and Jackknifing in Area Estimation Thomas P. Capizzi,
Hina Mehta,
Leonard Oppenheimer		 proc calis, auc, regression analysis source 6
95 Three Approaches of Developing Summary Software in SAS Frances Wilson proc step, proc means source 2
96 Towards an Interactive Statistical Analysis System Harold W. Gugel source
97 Two-way Analysis of Variance in Computer Statistical Packages: A Comparison of SPSS, BMDP, and RUMMAGE to SAS GLM David W. Garton,
Kenneth L. Koonce		 anova, main effect, proc glm, proc genmod, sample, data science, proc freq source 6
98 Type I Error Rates for the Unequal Variance T Test J. Philip Miller,
Robert P. Parks,
Reimut Wette		 sample, type i error, sample size, significance level, p-value, degrees of freedom, welch t-test, heteroscedasticity, f-test, student t distribution source
99 Using CMS SAS in Conjunction with IBM's System R Relational Data Base Management System J. P. Maile,
T. J. Vidmar		 sysrc, sample source 4
100 Using PROC G3D for Visualization of the Contour Plot of Residual Sums of Squares in PROC NLIN Derrick J. Bates The first step in solving a nonlinear equation is to provide some initial parameter estimates and then iterate from these to a final solution. However. without prior infor mation~ these initial estimates may be hard to produce. In addition there is always the danger of iterating to a local solution. One way of obtaining help for these problems is to set up a grid of parameter estimates, calculate residual sums of squares (RSS) for each point and search the resulting surface. both for possible starting values and local solutions. For two parameter nonlinear equations PROe NLIN has the capability of calculating and printing the RSS for each point on the grid and produc ing a line printer contour plot~ This paper presents a methodology for getting the grid points into a SAS data set and a scaling tech nique that has proven useful for subsequent three-dimensional plots by PROC G3D estimate, proc step source 4
101 Using SAS and the Data 100 Keybatch System in Establishing a SAS Data Base Lynn B. Eggering Setting up SAS Data Bases is tedious when there are numerous studies, record types, and variables. To make this prOCeSs more automatic and eliminate duplication of work efforts, we use the programm~ng featur~s in our data entry system (DATA 100 KEYBATC~ SYSTEM) and SAS to create our SAS files. clean and update our data, manage tape libraries, and generate reports. Using S~ col1ectixely and creatively with the DATA 100 KEYBATC~ System has proven beneficial to our data management requirements and our data processing time. INTRODOCTlON Essential to any project where there is a large amount of data to be entered, stored, and analysed, is a data entry system and a data management system that enhance the projects pro duction. We have accomplis~d this by using the software within the DATA 100 KEYBATC~ System, in conjuction with the flexability within SAS . to. streamline the creation of SAS files J clean and update our data, manage tape libraries, and generate reports macro definition, input statement source 2
102 Using SAS to Evaluate an Area Sampling Frame for Agricultural Surveys Ron Fesco,
Jeff Geuder,
Van Johnson		 Area frame surveys are a major source of data for crop and livestock estimates made by the Economics and Statistics Service (ESS). Frame construction errors can result in estimator biases, reduce sample efficiency, or both. To evaluate the quality of the frame, an Area Frame Analysis Package was developed which uses the Statistical Analysis System~ The analysis package is intended to be general enough to use for most ESS frames, and complete enough to require minimal programming. This paper briefly describes the area frame sampling methodology used by ESS and some problems encountered in frame construction. Examples of some modules in the package are included~ along with a brief interpretation of the output sample, proc step source 6
103 Using SAS to Manage A Large Clinical Trial Study William M. Taylor clinical trial, audit trail source
104 Using SAS to Reduce COMTEN Hardware Monitor Data Roger Berry,
Mary Sue Dickerson		 format statement, dataset source 6
105 Using SAS with the IBM 3800 and 6670 James D. Oleksiw SAS users who have access to an IBM 3800 laser printer or an tBM 6670 Information Distributor (Word P'rocessor) can make use of many of the inherent features of these devices. Very few changes to their existing programming methods are required. This paper examines some of the features of these devices and demonstrates how these capabilities can be incorporated in SAS programs proc step, sample, sas output source 7
106 Using Spatial Statistics to Produce Risk Qualified Maps R. A. Bilonick sample, correlation, covariance source
107 Using the SIR Data Management System as a Complement to SAS Jonathan Arlook missing value, statistical software, data governance, dataset source
108 Utilization of SAS Software for Market-Oriented Data Manipulation James C. Becknell,
Carl M. Peters,
Michele K. Steigleder		 batch processing, proc means source 5
109 Variance Estimates from Complex Survey Data: The Balanced Half-Sample Replication Procedure Josefina A. Lago sample, missing value source 5

Back to Conf go 1982 go 1980

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.