Multidimensional two-component Gaussian mixtures detection

Abstract

Let $(X_{1},\ldots,X_{n})$ be a $d$-dimensional i.i.d. sample from a distribution with density $f$. The problem of detection of a two-component mixture is considered. Our aim is to decide whether $f$ is the density of a standard Gaussian random $d$-vector ($f=\phi_{d}$) against $f$ is a two-component mixture: $f=(1-\varepsilon)\phi_{d}+\varepsilon\phi_{d}(\cdot -\mu)$ where $(\varepsilon,\mu)$ are unknown parameters. Optimal separation conditions on $\varepsilon$, $\mu$, $n$ and the dimension $d$ are established, allowing to separate both hypotheses with prescribed errors. Several testing procedures are proposed and two alternative subsets are considered.

Publication
Annales de l’Institut Henri Poincar'e, Probabilit'es et Statistiques (Série B), 54(2), pp. 842–865
Cathy MAUGIS-RABUSSEAU
Cathy MAUGIS-RABUSSEAU
Associate Professor