Open Conference Systems, ITACOSM 2019 - Survey and Data Science

Font Size: 
Borrowing strength from larger surveys to improve related estimates from smaller surveys using bivariate small area estimation models
Carolina Franco, William R Bell

Building: Learning Center Morgagni
Room: Aula Magna 327
Date: 2019-06-05 04:20 PM – 06:00 PM
Last modified: 2019-05-23


We demonstrate use of bivariate area-level models to improve small area estimates from one survey by borrowing strength from related estimates from a larger survey.  In particular, we demonstrate the potential for borrowing strength from estimates from the American Community Survey (ACS), the largest U.S. Household Survey, to imrpove estimates from smaller U.S. surveys.  For illustation we use, in conjunction with data from ACS, data from the National Health Interview Survey, the Survey of Income and Program Participation, and the Current Population Survey.  To borrow information we propose the use of a simple bivariate Guassian model and also, for proportions, a bivariate binomial logit normal model.  Simple theoretical calculations and the results from the examples show that substantial reductions in variances may be achieved by borrowing strength from the ACS via the bivariate models even without using regression covariates obtained form auxiliary sources.  Theoretical calculations how how the extent of variance reduction depends on the charactersitics of the underlying data.

Full Text: Slides