Multidimensional two-component Gaussian mixtures detection

Abstract
Let be a -dimensional i.i.d sample from a distribution with density . The problem of detection of a two-component mixture is considered. Our aim is to decide whether is the density of a standard Gaussian random -vector () against is a two-component mixture: where are unknown parameters. Optimal separation conditions on and the dimension are established, allowing to separate both hypotheses with prescribed errors. Several testing procedures are proposed and two alternative subsets are considered.
View on arXivComments on this paper