Objective Bayesian analysis for a spatial model with nugget effects

详细信息	查看全文 | 推荐本文 |

• 作者：Cuirong Ren^a ; ^{cuirong.ren@sdstate.edu} ; Dongchu Sun^b ; Chong He^b
• 关键词：Nugget effect ; Jeffreys prior ; Reference prior ; Integrated likelihood ; Propriety of posterior ; Frequentist coverage probability
• 刊名：Journal of Statistical Planning and Inference
• 出版年：2012
• 期刊代码：36_03783758
• 类别：mt
• 出版时间：July, 2012
• 卷：142
• 期：7
• 页码：1933-1946
• 文件大小：258 K

摘要

The focus of this paper is objective priors for spatially correlated data with nugget effects. In addition to the Jeffreys priors and commonly used reference priors, two types of 鈥渆xact鈥?reference priors are derived based on improper marginal likelihoods. An 鈥渆quivalence鈥?theorem is developed in the sense that the expectation of any function of the score functions of the marginal likelihood function can be taken under marginal likelihoods. Interestingly, these two types of reference priors are identical.

The propriety of the marginal priors and joint posteriors is studied for a large class of objective priors including all objective priors developed. Under quite general conditions, only Jeffreys-rule and 鈥渆xact鈥?reference priors yield proper posteriors. A simulation study shows that the exact reference priors perform better than the Jeffreys-rule prior. Two real spatial datasets are used for illustration.