The term used for the estimation of the misclassification rate in the discriminant analysis. Number of techniques has been proposed for two-group situation, but the multiple-group situation has rarely been addressed. The easiest procedure is the resubstitution technique, in which the training data are classified using the estimated classification rule and proportion incorrectly placed used as the estimate of misclassification rate. This technique is known to have a large optimistic bias, but it has the benefit that it can be applied to the multigroup problems with no modification required. An alternative technique is the leave one out estimator, in which each of the observation in turn is removed from the data and the classification rule recomputed using remaining data. The proportion improperly classified by the procedure will have reduced bias compared to resubstitution technique. This method can also be implied to the multi group problem with no modification but it has the large amount of variance.