This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 650
Type: Invited
Date/Time: Thursday, August 5, 2010 : 10:30 AM to 12:20 PM
Sponsor: Section on Survey Research Methods
Abstract - #306122
Title: An Empirical Bayes Approach for the National Agricultural Statistics Service's County Estimation Program
Author(s): Michael E. Bellow*+ and Parthasarathi Lahiri
Companies: National Agricultural Statistics Service and University of Maryland
Address: 3251 Old Lee Highway, Fairfax, VA, 22030,
Keywords: Empirical Bayes methods ; bootstrap ; small area estimation
Abstract:

In this paper, we consider an empirical Bayes method for combining multiple data sources in producing end-of-year estimates of crop harvested yield at the county level in the US. The method employs an area level model consisting of two submodels - one for the sampling error of direct survey estimates and the other relating true yields to a set of county specific predictor variables. Unlike standard variance component estimation methods such as maximum likelihood and residual maximum likelihood, the new methodology ensures strictly positive consistent estimation of model variance. In order to produce uncertainty measures of the proposed estimators, we introduce a parametric bootstrap method that incorporates all sources of uncertainty. Results of a study evaluating the empirical Bayes method for soybean yield in seven midwestern states are presented.


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 2010 program




2010 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.