When data come from a mixture of two weakly symmetric distributions with different means and proportional covariances, we estimate the best linear discriminant projection with the projection of the standardized data onto the direction of a vector-valued measure of skewness.

Weibull mixtures have been used extensively in reliability and survival analysis, and they have also been generalized by allowing negative mixing weights, which arise naturally under the formation of some structures of reliability systems. These models provide flexible distributions for modeling...