A method for the Generate a random sample from a finite mixture distributions

Dariush Ghorbanzadeh (1) , Philippe Durand (2) , Luan Jaupi (3)

1 CEDRIC - LAETITIA - CEDRIC. Traitement du signal et architectures électroniques
2 M2N - Modélisation mathématique et numérique
3 CEDRIC - MSDMA - CEDRIC. Méthodes statistiques de data-mining et apprentissage

A finite m ixture m odel i s a c onvex c ombination of more probability density functions. By combining the properties of the individual probability density functions, mixture models are capable of approximating any arbitrary distribution. In this work we propose a method for the Generate a random sample from a finite mixture distribution. The proposed method envelope conventional models: translation, scaling and translation-scaling.

Mots clés: Finite mixture distribution; Mixture of normal distri- butions; Skew normal distribution

Domaines: Statistiques [stat]

Dariush Ghorbanzadeh, Philippe Durand, Luan Jaupi. A method for the Generate a random sample from a finite mixture distributions. CMCGS 2017. 6th Annual International Conference on Computational Mathematics, Computational Geometry & Statistics, Mar 2017, Singapore, Singapore. ⟨10.5176/2251-1911_CMCGS17.3⟩. ⟨hal-02465159⟩