I discuss diagnostic methods for discriminant analysis. The equivalenc
e with linear regression is noted and regression diagnostics are consi
dered. The leverage is a function of the linear discriminant function
and the Mahalanobis distance of the observation from the group mean. T
he distribution of this distance is approximately chi-square with degr
ees of freedom equal to the number of variables. Standard tests may be
used to examine the normality assumption. Some examples are given.