Publikationsansicht

Bayesian Geoadditive Seemingly Unrelated Regression (2007)

Abstract
Parametric seemingly unrelated (SUR) models are... In this paper, we develop a Bayesian semiparametric SUR model, where the usual linear predictors are replaced by more flexible additive predictors allowing for simultaneous nonparametric estimation of such covariate effects and of spatial effects. The approach is based on appropriate smoothness priors which allow different forms and degrees of smoothness in a general framework. Inference is fully Bayesian and uses recent Markov chain Monte Carlo techniques.

Details der Publikation
Download http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.8.3966
Quelle http://www.stat.uni-muenchen.de/~lang/skript/sur_compstat_bw.pdf
Mitarbeiter CiteSeerX
Archiv CiteSeerX - Scientific Literature Digital Library and Search Engine (United States)
Keywords Bayesian semiparametric models, correlated responses, Markov random fields, MCMC, P-splines
Typ text
Sprache Englisch
Verknüpfungen 10.1.1.46.8136, 10.1.1.42.6990, 10.1.1.35.9617, 10.1.1.11.1480, 10.1.1.23.8918, 10.1.1.23.5270, 10.1.1.36.359