10
5

Multidimensional two-component Gaussian mixtures detection

Abstract

Let (X_1,,X_n)(X\_1,\ldots,X\_n) be a dd-dimensional i.i.d sample from a distribution with density ff. The problem of detection of a two-component mixture is considered. Our aim is to decide whether ff is the density of a standard Gaussian random dd-vector (f=ϕ_df=\phi\_d) against ff is a two-component mixture: f=(1ε)ϕ_d+εϕ_d(.μ)f=(1-\varepsilon)\phi\_d +\varepsilon \phi\_d (.-\mu) where (ε,μ)(\varepsilon,\mu) are unknown parameters. Optimal separation conditions on ε,μ,n\varepsilon, \mu, n and the dimension dd are established, allowing to separate both hypotheses with prescribed errors. Several testing procedures are proposed and two alternative subsets are considered.

View on arXiv
Comments on this paper