Their study concluded that multiple discriminant analysis
and recursive partitioning using decision tree gave the highest prediction accuracy for failed firms.
To perform the multiple discriminant analysis
of values and dimensions of organizational culture, we tested the assumptions of multivariate normality (using the Kolmogorov-Smirnov, or KS, test) and absence of outliers and homogeneity of the variance and covariance matrices (using Box's M test).
He used multiple discriminant analysis
to find a weighting (the coefficients) and combination of financial variables that would correctly classify the firms into their respective credit risk categories.
The main compositional approaches to the development of product spaces include common factor analysis (FA), principal components analysis (PCA) and multiple discriminant analysis
For discussions of multiple discriminant analysis
and logit, which have been used extensively in previous insolvency studies, see BarNiv and Hershbarger (1990).
Multiple discriminant analysis
(MDA) provides a procedure for assigning firms to predetermined groupings on the basis of variables whose values may depend on the group to which the firm actually belongs.
A summary of the step-wise multiple discriminant analysis
is shown in Table II.
A stepwise multiple discriminant analysis
was computed in order to determine value and importance of these variables for discriminating among the three employment status groups.
Bias in Multiple Discriminant Analysis
," Journal of Marketing Research, August 1965, 250-258.
The traditional approach and present standard for predicting financial distress uses multiple discriminant analysis
(MDA) to weight the relative value of information provided by a combination of financial ratios.
Chief among these methods have been multiple regression analysis, multiple discriminant analysis
and gravity models.
A multiple discriminant analysis
is a statistical procedure that is frequently used to distinguish (discriminate) between two or more populations on the basis of observations on several variables.