OPTIMAL TESTS FOR NO CONTAMINATION IN SYMMETRICAL MULTIVARIATE NORMALMIXTURES

SenGupta and Pal (1991, J. Statist. Plann. Inference, 29, 145-155) hav e recently obtained the locally optimal test for zero intraclass corre lation coefficient in symmetric multivariate normal mixtures, with kno wn mixing proportion, for the case when the common mean, m, and the co mmon variance, sigma2, are known. Here, we establish that even under t he general situation, when some or none of m and sigma2 are known, sim ple optimal tests can be derived, which are locally most powerful simi lar, whose exact cut-off points are already available and which retain all the previous optimality properties, e.g. unbiasedness, monotonici ty and consistency. Some power tables are presented to demonstrate the favorable performances of these tests.