Monte Carlo approximation of likelihood function in spatial GLMMs through an empirical Bayes method

ABSTRACT In spatial generalized linear mixed models (SGLMMs), statistical inference encounters problems, since random effects in the model imply high-dimensional integrals to calculate the marginal likelihood function. In this article, we temporarily treat parameters as random variables and express the marginal likelihood function as a posterior expectation. Hence, the marginal likelihood function is approximated using the obtained samples from the posterior density of the latent variables and parameters given the data. However, in this setting, misspecification of prior distribution of correlation function parameter and problems associated with convergence of Markov chain Monte Carlo (MCMC) methods could have an unpleasant influence on the likelihood approximation. To avoid these challenges, we utilize an empirical Bayes approach to estimate prior hyperparameters. We also use a computationally efficient hybrid algorithm by combining inverse Bayes formula (IBF) and Gibbs sampler procedures. A simulation study is conducted to assess the performance of our method. Finally, we illustrate the method applying a dataset of standard penetration test of soil in an area in south of Iran.

[1]  O. F. Christensen.  Monte Carlo Maximum Likelihood in Model-Based Geostatistics , 2004 .

[2]  Firoozeh Rivaz,et al.  Empirical Bayes spatial prediction using a Monte Carlo EM algorithm , 2009, Stat. Methods Appl..

[3]  Øivind Skare,et al.  Pairwise likelihood inference in spatial generalized linear mixed models , 2005, Comput. Stat. Data Anal..

[4]  Majid Jafari Khaledi,et al.  Non-Gaussian modeling of spatial data using scale mixing of a unified skew Gaussian process , 2013, J. Multivar. Anal..

[5]  Anthony Y. C. Kuk.  Automatic choice of driving values in Monte Carlo likelihood approximation via posterior simulations , 2003, Stat. Comput..

[6]  P. Diggle,et al.  Model‐based geostatistics , 2007 .

[7]  Tsai-Hung Fan,et al.  Bayesian analysis of multivariate t linear mixed models using a combination of IBF and Gibbs samplers , 2012, J. Multivar. Anal..

[8]  J. Berger,et al.  Objective Bayesian Analysis of Spatially Correlated Data , 2001 .

[9]  Hao Zhang.  Optimal Interpolation and the Appropriateness of Cross-Validating Variogram in Spatial Generalized Linear Mixed Models , 2003 .

[10]  T. C. Haas,et al.  Model-based geostatistics - Discussion , 1998 .

[11]  M. J. Khaledi,et al.  Bayesian spatial prediction of skew and censored data via a hybrid algorithm , 2015 .

[12]  G. Tian,et al.  A NONITERATIVE SAMPLING METHOD FOR COMPUTING POSTERIORS IN THE STRUCTURE OF EM-TYPE ALGORITHMS , 2003 .

[13]  D. Rubin.  Using the SIR algorithm to simulate posterior distributions , 1988 .

[14]  Anthony Y. C. Kuk,et al.  Monte Carlo approximation through Gibbs output in generalized linear mixed models , 2005 .

[15]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[16]  Gareth O. Roberts,et al.  Robust Markov chain Monte Carlo Methods for Spatial Generalized Linear Mixed Models , 2006 .

[17]  R. Waagepetersen,et al.  Bayesian Prediction of Spatial Count Data Using Generalized Linear Mixed Models , 2002, Biometrics.

[18]  S. T. Boris Choy,et al.  Nonignorable dropout models for longitudinal binary data with random effects: An application of Monte Carlo approximation through the Gibbs output , 2009, Comput. Stat. Data Anal..

[19]  H. Rue,et al.  Approximate Bayesian Inference in Spatial Generalized Linear Mixed Models , 2008 .

[20]  Hao Zhang.  On Estimation and Prediction for Spatial Generalized Linear Mixed Models , 2002, Biometrics.

[21]  G. Tian,et al.  Bayesian Missing Data Problems: EM, Data Augmentation and Noniterative Computation , 2009 .

[22]  P. Donnelly,et al.  Inference in molecular population genetics , 2000 .