Provides an interpretation of "discriminant analysis" that includes two sets of techniques, predictive discriminant analysis and descriptive discriminant analysis. Discusses basic purposes of these two analyses. Reviews common introductory practices associated with current applications of predictive and descriptive discriminant analysis. (Author)

The objective of this paper is to review the outputs of selected computer programs often used to carry out a "discriminant analysis" with respect to two purposes of such an analysis, discrimination and classification. The programs selected are three BMD programs. (Author/JKS)

Studied problems in reporting the use of predictive and descriptive discriminant analyses through an examination of 20 such analyses published from 1998 to 2000. Findings show that results were often very mixed and study purposes were typically not very explicit. (SLD)

An empirical comparison of three proposed indices of predictor variable potency is presented. These indices are: (1) the scaled weights of the first Fisher-type discriminant function, (2) the total group estimates of the correlations between each predictor variable and the first Fisher-type function, and (3) the within-groups estimates of the…

The purpose of this study was to determine which of six methods of ordering variables in a discriminant analysis yields subsets of variables that have the greatest discriminatory power. One method is based on univariate mean-square (or F) ratios, a second method on stepwise ordering, two methods on linear discriminant function (LDF) variable…

The process of discriminant analysis is reviewed in four different aspects: focusing on formulations, interpretations, and uses of discrimination, estimation and classification. Other issues, problems, and recent developments in discriminant analysis are also reviewed. (Author/DEP)

A subset of a real data set was used to illustrate an ad hoc analysis with missing data on multiple response variables. This strategy was initiated with a complete-case analysis to determine some variables that may be deleted with no loss in effects of interest. (SLD)

Predictive discriminant analysis and descriptive discriminant analysis are described, and the use of three popular statistical packages to obtain computational results for each type of discriminant analysis is reviewed. Results from two Biomedical Computer Program (BMDP), four Statistical Analysis System, and two Statistical Package for the Social…

A particular strategy for investigating effects resulting from a multivariate analysis of variance (MANOVA) is proposed. The strategy involves multiple two-group multivariate analyses. The two groups result from considering multivariate pairwise group contrasts or multivariate complex group contrasts. Assuming a given two-group analysis yields…

Classification is a procedure through which individuals are classified as being members of a particular group based on a variety of independent variables. Two methods of makin such classifications are discussed; the quadratic method is seen to be superior to the linear under certain constraints. (JKS)

Three methods of transforming unordered categorical response variables are described: (1) analysis using dummy variables; (2) eigenanalysis of frequency patterns scaled relative to within-groups variance; (3) categorical variables analyzed separately with scale values generated so that the grouping variable and the categorical variable are…

An empirical comparison is made of three proposed indices of relative predictor variable contribution: (1) the scaled weights of the first discriminant function; (2) the total group estimates of the correlations between each predictor variable and the first function; and (3) the within-groups estimates of the correlations between each predictor…

It is shown that in the special case of just two criterion groups the predictor variables may be equivalently ordered (with respect to contribution to prediction or discrimination) by the univariate F-ratios and by estimates of the predictor versus the linear discriminant function correlations. (Author)

Some interpretations of relative variable importance in the contexts of multivariate analysis of variance (MANOVA) and discriminant analysis (DA) are presented. Some indices potentially useful for the interpretations are presented, and the assessment of variable importance is illustrated using real data sets. Both descriptive discriminant analysis…

There are two distinct but related purposes for carrying out a "discriminant analysis": (1) discrimination, and (2) classification. The primary objective of this paper was to review the outputs of selected computer programs often used to carry out a "discriminant analysis" with respect to these two purposes. Information provided by the programs on…