Yvik Swan, Université Libre de Bruxelles: Univariate and multivariate Chen-Stein characterizations - a parametric approach
Abstract: We provide a general framework for characterizing families of (univariate, multivariate, discrete and continuous) distributions in terms of a parameter of interest. We show how this allows for recovering known Chen-Stein characterizations, and for constructing many more. Several examples are worked out in full, and different potential applications are discussed.
Philippe Lambert, Université catholique de Louvain : Nonparametric additive models for interval-censored data
An additive model for the location and dispersion of a continuous response with an arbitrary smooth conditional distribution is proposed. B-splines are used to specify the three components of the model. It is extended to deal with an interval censored response and potentially interval censored covariates. Monte-Carlo Markov chains are generated to explore the joint posterior distribution of the spline coecients and of the penalty parameters controlling the smoothness of the functional components in the model. After an extensive simulation study, the model is fitted on datasets from human and medical sciences.
Cetinyurek, A. and Lambert, P. (2011) Smooth estimation of survival functions and hazard ratios from interval-censored data using Bayesian penalized B-splines. Statistics in Medicine, 30: 75-90.
Lambert, P. (2012) Nonparametric additive location-scale models for interval censored data. Statistics and Computing (in press).
Lambert, P. (2011) Smooth semi- and nonparametric Bayesian estimation of bivariate densities from bivariate histogram data. Computational Statistics and Data Analysis, 55: 429-445.
Lambert, P. and Eilers, P.H.C. (2009) Bayesian density estimation from grouped continuous data. Computational Statistics and Data Analysis, 53: 1388-1399.