Screening among multivariate normal data

This paper considers the problem of screening k multivariate normal populat ions (secondary data) with respect to a control population (primary data) i n terms of covariance structure. A screening procedure, developed based upo n statistical ranking and selection theory, is designed to include in the s elected subset those populations which have the same (or similar) covarianc e structure as the control population, and exclude those populations which differ significantly. Formulas for computing the probability of a correct s election and the least favorable configuration are developed. The sample si ze required to achieve a specific probability requirement is also developed , with results presented in tabular form. This secondary data selection pro cedure is illustrated via an example with applications to radar signal proc essing. (C) 1999 Academic Press.