A test for the mean vector in large dimension and small samples

Abstract In this paper, we consider the problem of testing the mean vector in the multivariate setting where the dimension p is greater than the sample size n, namely a large p and small n problem. We propose a new scalar transform invariant test and show the asymptotic null distribution and power of the proposed test under weaker conditions than Srivastava (2009) . We also present numerical studies including simulations and a real example of microarray data with comparison to existing tests developed for a large p and small n problem.