Activity Number:
|
255
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 1, 2016 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Survey Research Methods Section
|
Abstract #319235
|
View Presentation
|
Title:
|
A Design-Based Approach to Small-Area Estimation Using Semiparametric Generalized Linear Mixed Model
|
Author(s):
|
Hongjian Yu* and Yueyan Wang and Pan Wang and Jean Opsomer and Ninez Ponce and David Grant
|
Companies:
|
University of California at Los Angeles and University of California at Los Angeles and University of California at Los Angeles and and University of California at Los Angeles and University of California at Los Angeles
|
Keywords:
|
Small Area Estimation ;
Penalized Spline ;
Generalized Linear Mixed Model
|
Abstract:
|
Non-parametric model using penalized spline regression in small area estimation context was proposed by Opsomer et al (2008). Wang et al (2015) applied this technique in granular area estimations. It was shown to be a useful tool to provide supplemental information where survey observations are few or non-existent. In this paper we further examine semiparametric generalized linear mixed model in producing consistent estimations for multiple area levels. We use the mosaic analogy to describe the process. We employ a design-based Jackknife method for variance calculation.
|
Authors who are presenting talks have a * after their name.