An estimate for the loss of the classification power when a single variate is excluded from the linear discriminant function is given. A criterion is proposed for the omission of a variate without an ...

LDA is based on the extraction of linear discriminant functions of the independent variables by means of a qualitative dependent variable and several quantitative independent variables.

Asymptotic expansions of the distributions of the linear and the quadratic discriminant functions are derived. The expected misclassification probabilities of these two functions are compared when the ...