AN ALTERNATIVE PROOF FOR A KNOWN RESULT OF NONCENTRAL WISHART DISTRIBUTION

In the theory of multivariate statistics, it is well known that given a sample of n independent p-variate normally distributed random vector s with a common variance covariance matrix, if at least one of the n v ectors has nonzero means, then the sum of squares about the sample mea n of the n vectors has a noncentral Wishart distribution. However, a d etailed proof for this known result is rarely found in literature. In this paper, we present a formal and complete proof for the well-known result together with an example of its applications.