[Home ] [Archive]   [ فارسی ]  
:: Main :: About :: Current Issue :: Archive :: Search :: Submit :: Contact ::
:: Volume 10, Issue 2 (3-2017) ::
J. of Stat. Sci. 2017, 10(2): 233-260 Back to browse issues page
Approximate Bayesian Analysis of Spatio-Temporal Data Using a Gaussian Markov Random Field
Fatemeh Hosseini , Elham Homayonfal
Abstract:   (4461 Views)

Hierarchical spatio-temporal models are used for modeling space-time responses and temporally and spatially correlations of the data is considered via Gaussian latent random field with Matérn covariance function. The most important interest in these models is estimation of the model parameters and the latent variables, and is predict of the response variables at new locations and times. In this paper, to analyze these models, the Bayesian approach is presented. Because of the complexity of the posterior distributions and the full conditional distributions of these models and the use of Monte Carlo samples in a Bayesian analysis, the computation time is too long. For solving this problem, Gaussian latent random field with Matern covariance function are represented as a Gaussian Markov Random Field (GMRF) through the Stochastic Partial Differential Equations (SPDE) approach. Approximatin Baysian method and Integrated Nested Laplace Approximation (INLA) are used to obtain an approximation of the posterior distributions and to inference about the model. Finally, the presented methods are applied to a case study on rainfall data observed in the weather stations of Semnan in 2013.

Keywords: Spatio-Temporal Data, Gaussian Markov Random Field, Integrated Nested Laplace Approximation, Stochastic Partial Differential Equations.
Full-Text [PDF 752 kb]   (1215 Downloads)    
Type of Study: Applied | Subject: Spatial Statistics
Received: 2015/01/23 | Accepted: 2015/11/9 | Published: 2016/12/20
Send email to the article author

Add your comments about this article
Your username or Email:

Write the security code in the box >



XML   Persian Abstract   Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

Hosseini F, Homayonfal E. Approximate Bayesian Analysis of Spatio-Temporal Data Using a Gaussian Markov Random Field. J. of Stat. Sci.. 2017; 10 (2) :233-260
URL: http://jss.irstat.ir/article-1-352-en.html


Volume 10, Issue 2 (3-2017) Back to browse issues page
مجله علوم آماری – نشریه علمی پژوهشی انجمن آمار ایران Journal of Statistical Sciences
Persian site map - English site map - Created in 0.05 seconds with 31 queries by YEKTAWEB 3647