On p/n-asymptoticsapplied to traces of 1st and 2nd order powers of Wishart matrices with application to goodness-of-fit testing
(English)Manuscript (preprint) (Other academic)
The distribution of the vector of the normalized traces of and of , where the matrix follows a matrix normal distribution and is proved, under the Kolmogorov condition , to be multivariate normally distributed. Asymptotic moments and cumulants are obtained using a recursive formula derived in Pielaszkiewicz et al. (2015). We use this result to test for identity of the covariance matrix using a goodness–of–fit approach. The test performs well regarding the power compared to presented alternatives, for both c < 1 or c ≥ 1.
goodness–of–fit test, covariance matrix, Wishart matrix, multivariate normal distribution
IdentifiersURN: urn:nbn:se:liu:diva-122620OAI: oai:DiVA.org:liu-122620DiVA: diva2:868934