Author: Jae-kwang Kim (Department of Statistics, Iowa State University, Ames, IA)
Combining information from two independence surveys with similar measurement can be a promising area of research in small area estimation. To incorporate the survey specific effect, we use a random effect model in the population level. The sampling design can be informative in the sense that the sample distribution can be different from that of the population. A pretest method can be developed to test the significant of the random effect. We further develop a shrinkage estimation for the variance component in the random effect model. The proposed method is applied to a real data example in Korean surveys.
Only logged in users can see slides when author's permission was given. Please register to have access.