Smooth semi- and nonparametric bayesian estimation of bivariate densities from bivariate histogram data
(2009) , 29 pages
Files
Smoothsemi-andnonparametricbayesianestimationofbivariatedensitiesfrombivariatehistogramdata.pdf
Open Access - Adobe PDF
- 820.87 KB
Details
- Authors
- Abstract
- We show how penalized B-splines combined with the composite link model can be used to estimate a bivariate density from histogram data. Two strategies are proposed: the first one is semi-parametric with exible margins modeled using B-splines and a parametric copula for the dependence structure ; the second one is nonparametric and is based on Kronecker products of the marginal B-splines bases. Frequentist and Bayesian estimations are described. A large simulation study quantifies the performances of both methods under dierent dependence structures and varying strengths of dependence, sample sizes and amounts of grouping. It suggests that Schwarz's BIC is a good tool for classifying the competing models. The density estimates are used to evaluate conditional quantiles in two applications in social and in medical sciences.
- Affiliations
Uliège Institut des sciences humaines et sociales
Citations
Lambert, P. (2009). Smooth semi- and nonparametric bayesian estimation of bivariate densities from bivariate histogram data (Stat Discussion Paper 0935). https://hdl.handle.net/2078.5/35141