Regression; *Predictive Discriminant Analysis

A method for comparing the cross-validated classification accuracies of predictive discriminant analysis and logistic regression classification models is presented under varying data conditions for the two-group classification problem. With this method, separate-group, as well as total-sample proportions of the correct classifications, can be compared for the two models. The test for contrasting correlated proportions developed by Q. McNemar (1947) is used in the statistical comparisons of the separate-group data and total-sample proportions. The method is illustrated with 32 real data sets that varied in number of cases, relative group sizes, number of predictor variables, degree of group separation, and equality of group covariance matrices. (Contains 1 table and 24 references.) (SLD) ***.A**-A**** * *.AA-*****,********************************** * Reproductions supplied by EDRS are the best that can be made from the original document