Mod\`ele \`a processus latent et algorithme EM pour la r\'egression non lin\'eaire
Statistics Theory
2013-12-30 v1 Machine Learning
Methodology
Machine Learning
Statistics Theory
Abstract
A non linear regression approach which consists of a specific regression model incorporating a latent process, allowing various polynomial regression models to be activated preferentially and smoothly, is introduced in this paper. The model parameters are estimated by maximum likelihood performed via a dedicated expecation-maximization (EM) algorithm. An experimental study using simulated and real data sets reveals good performances of the proposed approach.
Cite
@article{arxiv.1312.6978,
title = {Mod\`ele \`a processus latent et algorithme EM pour la r\'egression non lin\'eaire},
author = {Faicel Chamroukhi and Allou Samé and Gérard Govaert and Patrice Aknin},
journal= {arXiv preprint arXiv:1312.6978},
year = {2013}
}