A Forecast-oriented Parameter Estimation for Geostatistical Models
Dr. Hong-Ding Yang
The Graduate Institute of Statistics and Information Science National Changhua University of Education
I consider geostatistical regression models to predict spatial variables of interest. It is known that parameters in the Matérn covariogram cannot be estimated well by the likelihood-based methods. Although a best linear unbiased predictor has been proposed when model parameters are known, a predictor with estimated parameters is nonlinear and may be not the best in practice. To evaluate the predicted ability of the nonlinear spatial predictor, an unbiased L2-risk estimator with an adjusted procedure for the likelihood-based estimates is proposed. The adjusted parameter estimators based on minimizing a corrected Stein’s unbiased risk estimator tend to have less bias than the conventional likelihood-based estimators, and the resulting spatial predictor is more accurate and more stable. Statistical inference for the proposed methodology is justified both theoretically and numerically. To verify the practicability of the proposed method, a groundwater data set in Bangladesh is analyzed. Finally, some results regarding my current researches will be illustrated briefly.
Keywords: geostatistics, Matérn covariogram, parameter estimation, smoothing parameter, spatial prediction.
日 期：108年3月6日(星期三) 16:10~17:00
茶 會： 15:30~16:00數學館三樓305室舉行