CMStatistics 2016: Start Registration
View Submission - CMStatistics
B0723
Title: Copula-based estimation of conditional densities and hazard rate functions Authors:  Noel Veraverbeke - Hasselt University (Belgium) [presenting]
Paul Janssen - Hasselt University (Belgium)
Jan Swanepoel - North West University Potchefstroom Campus (South Africa)
Abstract: New copula-based smooth Bernstein estimators for the conditional density function and the related conditional hazard rate function are presented. The conditional density estimator is defined as a smoother of the copula-based Bernstein estimator of the conditional distribution function. In a similar way, the conditional hazard rate estimator is a smoother of the conditional cumulative hazard rate function estimator. We discuss the asymptotic properties of the bias, variance and distribution in terms of the smoothing parameters.