Interpolation and Extrapolation in Random Coefficient Regression Models: Optimal Design for Prediction
The problem of optimal design for the prediction of individual parameters in random coefficient regression in the particular case of a given population mean was considered by Gladitz and Pilz (Statistics 13:371–385, 1982). In the more general situation, where the population parameter is unknown, D- and L-optimal designs were discussed in Prus and Schwabe (J R Stat Soc Ser B, 78:175–191). Here we present analytical results for designs which are optimal for prediction in the case of interpolation as well as extrapolation of the individual response.
KeywordsFixed Effect Model Individual Parameter Error Matrix Dispersion Matrix Linear Unbiased Predictor
- 2.Chaloner, K.: Optimal Bayesian experimental design for linear models. Department of Statistics Technical report 238, Carnegie-Mellon University, Pittsburgh (1982)Google Scholar
- 9.Prus, M.: Optimal designs for the prediction in hierarchical random coefficient regression models. Ph.D. thesis, Otto-von-Guericke University Magdeburg (2015)Google Scholar
- 10.Prus, M., Schwabe, R.: Optimal designs for individual prediction in random coefficient regression models. In: Melas, V., Nachtmann, G., Rasch, D. (eds.) Optimal Design of Experiments – Theory and Application: Proceedings of the International Conference in Honor of the late Jagdish Srivastava, pp. 122–129. Center of Experimental Design, University of Natural Resources and Life Sciences, Vienna (2011)Google Scholar
- 11.Prus, M., Schwabe, R.: Optimal designs for the prediction of individual parameters in hierarchical models. J. R. Stat. Soc.: Ser. B. 78, 175–191 (2016)Google Scholar