|
Activity Number:
|
427
- Contributed Poster Presentations:Government Statistics Section
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Government Statistics Section
|
|
Abstract #330921
|
|
|
Title:
|
Bayesian Estimation with Shrinking Both Means and Variances in Heteroscedastic Nested Error Regression Models
|
|
Author(s):
|
Hiromasa Tamae*
|
|
Companies:
|
|
|
Keywords:
|
small-area estimation;
hierarchical Bayesian model;
random dispersion;
nested error regression model;
shrinkage prior
|
|
Abstract:
|
This research concerns small-area estimation in the heteroscedastic nested error re- gression (HNER) model which assumes that the within-area variances are different among areas. This model is useful for analyzing data such that the within-area variation changes from area to area, but direct estimates of the variances based on the within-area data are poor because of small samples sizes for small-areas. An approach to treating the problem is to consider the hierarchical Bayesian model as- suming the prior distributions for the area variances. The resulting Bayes estimator is stable because both means and variances of small areas are shrunken towards stable statistics.
|
Authors who are presenting talks have a * after their name.
