The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:

16

Type:

Topic Contributed

Date/Time:

Sunday, July 29, 2012 : 2:00 PM to 3:50 PM

Sponsor:

Section on Survey Research Methods

Abstract  #304255 
Title:

Inference for Finite Population Quantiles of NonNormal Survey Data Using Bayesian Mixture of Splines

Author(s):

Qixuan Chen*+ and Xuezhou Mao and Michael R. Elliott and Roderick Little

Companies:

Columbia University and Columbia University and University of Michigan and University of Michigan/U.S. Census Bureau

Address:

722 West 168th Street, New York, NY, , USA

Keywords:

Bayesian inference ;
Gibbs sampling ;
penalized spline regression ;
probabilityproportionaltosize sampling

Abstract:

Nonnormally distributed data are common in sample surveys. We propose a robust Bayesian modelbased estimator for finite population quantiles of nonnormal survey data in probabilityproportionaltosize sampling. We assume that the probability of inclusion is known for all the units in the finite population. The nonnormal distribution of the continuous survey variable is approximated using a mixture of normal distributions, in which both the mean and the variance of the survey variable are modeled as a smooth function of the inclusion probabilities in each mixture component. A full Bayesian approach using the Markov chain Monte Carlo method is developed to obtain the posterior distribution of the finite population quantiles. We compare our proposed estimator with alternative estimators using simulations based on artificial data as well as a real finite population.

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 2012 program

2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 2313473.
If you have questions about the Continuing Education program, please contact the Education Department.