|
Activity Number:
|
133
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #306815 |
|
Title:
|
Robust Estimation of the Mean Square Error of an EBLUP of a Small-Area Mean
|
|
Author(s):
|
Shijie Chen*+ and Partha Lahiri and Jon N. K. Rao
|
|
Companies:
|
RTI International and University of Maryland and Carleton University
|
|
Address:
|
701 13th Street, NW, Washington, DC, 20005-3967,
|
|
Keywords:
|
|
|
Abstract:
|
In this paper, we present a general method for estimating the mean square error (MSE) of EBLUP for the well-known Fay-Herriot small-area model. Unlike the normality-based MSE approximations, our result involves the kurtosis (but not the skewness) of both the sampling and model errors and depends on the method of estimating variance component. For the method of moments estimator of the variance component, estimation of the kurtosis of the model error is not required. Estimation of the kurtosis is necessary for other methods of estimating variance components (e.g., the method proposed by Fay and Herriot (1979)). We propose a method for estimating the kurtosis that provides a method of MSE estimation for a general class of variance component estimators with the non-normal sampling and model errors.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2006 program |
