PLACE:  296 Town Engineering

SPEAKER:
Tapabrata Maiti
University of Nebraska - Lincoln and  U. S. Census Bureau
 
TITLE:
Small Area Estimation Using Time Series and Cross-Sectional Data

ABSTRACT:

Small area estimation has received considerable importance in survey sampling
due to its growing demand from both public and private sectors. It is well
known that the direct survey estimates are likely to yield unacceptably large
standard errors due to the smallness of sample sizes in the areas.  The
appropriateness of model-based inference in small area estimation is well
recognized. Suitable models are used to produce reliable small area estimates
by `borrowing strength' from related auxiliary information.

In this article, a general methodology is proposed to combine information from
similar regions and past surveys. The method is directly applied to the
estimation of median income of four-person families for the fifty states and
the district of Columbia. The proposed method is an advancement over the method
currently used by the U.S. Bureau of the Census in the sense that it uses a
more reliable model and produces a valid measure of uncertainty of the proposed
estimates.
 
 

COFFEE: 10:30 a.m., 104 Snedecor