Title: Simplified estimations of conditional copulas
Authors: Jean-David Fermanian - Ensae-Crest (France) [presenting]
Abstract: Semiparametric conditional copula models suffer from the so-called curse of dimensionality. Indeed, conditional marginal distributions with a potentially large number of covariates have to be estimated with usual smoothing techniques. By assuming diverse single-index assumptions for such conditional distributions, we propose a simple way of reducing this curse of dimensionality. Therefore, with an underlying parametric conditional copula model under the simplifying assumption, some theoretical properties of estimated parameter are provided.