PROJECTION PURSUIT THROUGH RELATIVE ENTROPY MINIMISATION

Jacques Touboul (1)

1 LSTA - Laboratoire de Statistique Théorique et Appliquée

Projection Pursuit methodology permits to solve the difficult problem of finding an estimate of a density defined on a set of very large dimension. In his seminal article, Huber (see "Projection pursuit", Annals of Statistics, 1985) evidences the interest of the Projection Pursuit method thanks to the factorisation of a density into a Gaussian component and some residual density in a context of Kullback-Leibler divergence maximisation. In the present article, we introduce a new algorithm, and in particular a test for the factorisation of a density estimated from an iid sample.

Mots clés: Projection pursuit; Minimum Kullback-Leibler divergence maximisation; Robustness; Elliptical distribution

Domaines: Statistiques [math.ST]; Théorie [stat.TH]

Jacques Touboul. PROJECTION PURSUIT THROUGH RELATIVE ENTROPY MINIMISATION. 2010. ⟨hal-00509610⟩