Journal article
A comparative study of Gaussian geostatistical models and Gaussian Markov random field models
Journal of multivariate analysis, Vol.99(8), pp.1681-1697
09/2008
DOI: 10.1016/j.jmva.2008.01.012
PMCID: PMC2662683
PMID: 19337581
Abstract
Gaussian geostatistical models (GGMs) and Gaussian Markov random fields (GMRFs) are two distinct approaches commonly used in spatial models for modeling point-referenced and areal data, respectively. In this paper, the relations between GGMs and GMRFs are explored based on approximations of GMRFs by GGMs, and approximations of GGMs by GMRFs. Two new metrics of approximation are proposed : (i) the Kullback–Leibler discrepancy of spectral densities and (ii) the chi-squared distance between spectral densities. The distances between the spectral density functions of GGMs and GMRFs measured by these metrics are minimized to obtain the approximations of GGMs and GMRFs. The proposed methodologies are validated through several empirical studies. We compare the performance of our approach to other methods based on covariance functions, in terms of the average mean squared prediction error and also the computational time. A spatial analysis of a dataset on PM2.5 collected in California is presented to illustrate the proposed method.
Details
- Title: Subtitle
- A comparative study of Gaussian geostatistical models and Gaussian Markov random field models
- Creators
- Hae-Ryoung SongMontserrat Fuentes - North Carolina State UniversitySujit Ghosh
- Resource Type
- Journal article
- Publication Details
- Journal of multivariate analysis, Vol.99(8), pp.1681-1697
- DOI
- 10.1016/j.jmva.2008.01.012
- PMID
- 19337581
- PMCID
- PMC2662683
- NLM abbreviation
- J Multivar Anal
- ISSN
- 0047-259X
- eISSN
- 1095-7243
- Publisher
- Elsevier
- Grant note
- The research conducted by Fuentes has been partly supported by a National Science Foundation grants DMS 0353029 and DMS 0706731, the National Institutes of Health award R01 ES01 14884301A2, and the Environmental Protection Agency award R833863. Song has been supported by a cooperative agreement (CT829562) with the Office of Air Quality Planning and Standards/US Environmental Protection Agency.
- Language
- English
- Date published
- 09/2008
- Academic Unit
- Statistics and Actuarial Science; Biostatistics; Provost Office Administration
- Record Identifier
- 9983763497002771
Metrics
46 Record Views