B1246
Title: Kernel density estimation for continuous time processes with values in a Riemannian manifold
Authors: Anne Francoise Yao - Universite Clermont Auvergne/LMBP (France) [presenting]
Vincent Monsan - Universite Felix Houphouet Boigny - Laboratoire LAMI (Cote d'Ivoire)
Djack Guy-Aude Kouadio - Universite Felix Houphouet Boigny - Laboratoire LAMI (Cote d'Ivoire)
Abstract: The purpose is to address the problem of estimation of the density of the univariate marginal distribution of a strong mixing continuous time process. This topic has been widely treated in the literature in the case where the process is with values in an Euclidean space. However, the situation where the process lives in a Riemannian submanifold has yet to be studied. A prior study has proposed a kernel density estimator for independent data in the Riemannian submanifold. An integral counterpart of Pelletier's estimator is addressed for continuous time processes, and some related weak and strong consistency results are given.
