Title: Copula-based logistic regression estimation
Authors: Taoufik Bouezmarni - Universite de Sherbrooke (Canada) [presenting]
Abstract: A new methodology is explored for estimating the success probability in a logistic regression setting. The idea consist in writing this later in terms of conditional copula and marginal distributions. The approach developing here consist, first, selecting a parametric family of copula that describes better the data at hand, and estimating non parametrically the marginal distributions, subsequently, we use the plug-in method to build an estimator the desired probability either in a binary case or categorical one. The availability of a rich family of copula, a various goodness-of-fit test and a nonparametric estimation of the marginal makes the approach more flexible. The asymptotic properties related to this estimators are provided. Finally, a simulated study is carried out to evaluate the performance of the newly proposed procedure, whether on simulated examples or real data.