Communication dans un congrès
Titre :
Bayesian Selection of Adaptive Bandwidth in Non-homogeneous Poisson Process Kernel Estimators for the Intensity Function
Auteurs :
Marcel Sihintoé Badiane, Papa Ngom, Clément Manga
Résumé :
The fundamental problem in the kernel intensity estimation is the choice of the bandwidth. In this paper, the proposed estimator for the bivariate intensity function for non-homogeneous Poisson Process is based on bivariate continuous associated kernels. It is well known that, the serious problem inherent in this approach is that performance of the kernel estimator depends on the selection of a bandwidth parameter. To overcome the problem, we propose a Bayesian adaptive approach to select the vector of bandwidths considering the bivariate intensity function and the quadratic and entropy loss functions are considered. Exact formulas for the posterior distribution and the vector of bandwidths are obtained. Numerical studies indicate that the performance of our approach is better, comparing with other bandwidth selection techniques using integrated squared error as criterion. Some applications are made on real datasets.
Mots-clés :
Bayesian Analysis, Non-parametric inference, Non-homogeneous poisson process
Congrès :
Second International Symposium of Non-Linear Analysis Geometry and Applications
Date :
26-29 Janvier 2022
Ville :
Ziguinchor
Pays :
Senegal
Pages :
143-169
Date de publication :
April 2022
Lien de la publication :
Voir
ici
événements