|
Activity Number:
|
427
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #310130 |
|
Title:
|
Parametric Bootstrap Confidence Interval in a Small-Area Estimation Problem
|
|
Author(s):
|
Huilin Li*+ and Partha Lahiri
|
|
Companies:
|
University of Maryland and University of Maryland
|
|
Address:
|
Rm. 4315 Math Building, College Park, MD, 20740,
|
|
Keywords:
|
Parametric Bootstrap ; confidence interval ; small area estimation ; Fay-Herriot model ; ADM estimator
|
|
Abstract:
|
In this paper, we developed the theory of using ADM variance estimator in a parametric bootstrap method in constructing confidence intervals of small area means. Using a Monte Carlo simulation study, we first investigate the performance of different variance component methods in the parametric bootstrap confidence interval, and then we compare our method with different rival methods in terms of coverage probabilities and average lengths. We then demonstrate the utility of the parametric bootstrap method by analyzing several real life datasets.
|
