Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
1144936 | Journal of the Korean Statistical Society | 2009 | 6 Pages |
Abstract
For spatially correlated repeated arrays, a simple method is proposed for maximum likelihood (ML) estimation of the mean parameters. Efficiency of the sample mean over the maximum likelihood estimator (MLE) is analyzed. Spatial correlations combined with heterogeneity of spatial correlations or heterogeneity of error variances are shown to have adverse effect on efficiency of the sample mean. Therefore, in such spatially correlated and heterogeneous situations, it is recommended that spatial correlations should be properly addressed in estimating mean parameters.
Related Topics
Physical Sciences and Engineering
Mathematics
Statistics and Probability
Authors
Dong Wan Shin,