Computation of Empirical Bayes Estimates Using Single Level Mixed Models
KEY WORDS: Empirical Bayes, Mixed Models
This paper is a description of the algorithms and code that were developed in R
to compute empirical Bayes estimates. The setting is that some variables that have direct survey estimates for some set of geographic areas or domains are given along with estimates
of the variance of those variables. In addition, a set of variables that are
correlated with the variable of interest is given. An empirical Bayes estimate is then
a weighted average of the direct estimate and a regression estimate. It turns out that
the empirical Bayes method in this setting is equivalent to a single level mixed model
with known variances.
CITATION: Creecy, Robert. "Computation of Empirical Bayes Estimates Using Single Level Mixed Models."
Source: U.S. Census Bureau, Statistical Research Division
Created: August 5, 2008
Last revised: August 4, 2008
This symbol indicates a link to a non-government web site. Our linking to these sites does not constitute an endorsement of any products, services or the information found on them. Once you link to another site you are subject to the policies of the new site.