Bayesian Geoadditive Seemingly Unrelated Regression 1

Parametric seemingly unrelated regression (SUR) models are a common tool for multivariate regression analysis when error variables are reasonably correlated, so that separate univariate analysis may result in inefficient estimates of covariate effects. A weakness of parametric models is that they require strong assumptions on the functional form of possibly nonlinear effects of metrical covariates. 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.

[1]  S. Lang,et al.  Bayesian P-Splines , 2004 .

[2]  Rick Chappell,et al.  Multivariate Statistical Modelling Based on Generalized Linear Models , 2003 .

[3]  Dani Gamerman,et al.  Space-varying regression models: specifications and simulation , 2001, Comput. Stat. Data Anal..

[4]  H. Rue,et al.  On Block Updating in Markov Random Field Models for Disease Mapping , 2002 .

[5]  Eric R. Ziegel,et al.  Multivariate Statistical Modelling Based on Generalized Linear Models , 2002, Technometrics.

[6]  Harald Hruschka,et al.  Market Share Analysis Using Semi-Parametric Attraction Models , 2001, Eur. J. Oper. Res..

[7]  L. Fahrmeir,et al.  Bayesian Semiparametric Regression Analysis of Multicategorical Time-Space Data , 2001 .

[8]  L. Fahrmeir,et al.  Bayesian inference for generalized additive mixed models based on Markov random field priors , 2001 .

[9]  Ludwig Fahrmeir,et al.  Semiparametric Analysis of the Socio-Demographic and Spatial Determinants of Undernutrition in Two African Countries , 2001 .

[10]  R. Kohn,et al.  Nonparametric seemingly unrelated regression , 2000 .

[11]  R. Carroll,et al.  Nonparametric Function Estimation for Clustered Data When the Predictor is Measured without/with Error , 2000 .

[12]  D. Dey,et al.  Bayesian analysis for correlated Ordinal Data Models , 2000 .

[13]  H. Rue.  Fast Sampling of Gaussian Markov Random Fields with Applications , 2000 .

[14]  Håvard Rue,et al.  On block updating in Markov random field models for disease mapping. (REVISED, May 2001) , 2000 .

[15]  Harald J. van Heerde,et al.  Semiparametric Analysis to Estimate the Deal Effect Curve , 2001 .

[16]  M. Pourahmadi.  Joint mean-covariance models with applications to longitudinal data: Unconstrained parameterisation , 1999 .

[17]  A. Montgomery.  Creating Micro-Marketing Pricing Strategies Using Supermarket Scanner Data , 1997 .

[18]  Thomas W. Yee,et al.  Additive extensions to generalized estimating equation methods , 1996 .

[19]  Jianqing Fan,et al.  Local polynomial modelling and its applications , 1994 .

[20]  B. Silverman,et al.  Nonparametric Regression and Generalized Linear Models: A roughness penalty approach , 1993 .

[21]  J. Besag,et al.  Bayesian image restoration, with two applications in spatial statistics , 1991 .

[22]  J. Pasciak,et al.  Computer solution of large sparse positive definite systems , 1982 .

[23]  C. R. Deboor,et al.  A practical guide to splines , 1978 .