Bayesian pattern mixture model under nonignorable nonresponse for binary data
Sukyoung An 1 · Balgobin Nandram 2 · Dal Ho Kim 3
13 Department of Statistics, Kyungpook National University
2 Department of Mathematical Sciences, Worcester Polytechnic Institute
Received 24 June 2019, revised 12 July 2019, accepted 12 July 2019
Abstract
We consider a Bayesian pattern mixture model to estimate the proportion of the finite population with missing data. The pattern mixture approach is a way to model missing data. We describe the Bayesian model considering two cases for the parameter of a prior distribution. To fit the model, we use Markov chain Monte Carlo methods.
We use the Gibbs sampler with grid method to get the samples of the parameters.
We use the National Crime Survey data summarized by Stasny (1991) to estimate the proportion of the finite population. When considering two cases of the parameter of a prior distribution, we saw that the inference for the parameter was not sensitive in our proposed model.
Keywords: Bayesian estimation, grid methods, latent variable, pattern mixture model.
1. Introduction
Many data are collected to identify what is happening in society. In the data collection process, missing data are included for a variety of reasons. If you don’t know the information about missing data, the inference for the data becomes difficult and you get inaccurate conclusions. Missing data can be ignored if there is no significant information, but data should be taken into account if missing data has significant information. Many studies have been done from the past to deal with missing data. To make inference or estimation using missing data, we should have various background knowledge such as understanding the missing data mechanism, considering the missing data model, and the methods for analysis model. Ma and Chen (2018) reviewed developments and applications for handling missing data in Bayesian methods.
The missing data mechanism is divided into three types (Little and Rubin, 2002). When the missing values do not depend on the missing or observed data, the values are called missing
1
Ph.D. candidate, Department of Statistics, Kyungpook National University, Daegu 41566, Korea.
2
Professor, Department of Mathematical Sciences, Worcester Polytechnic Institute, 100 Institute Road, Worcester, MA 01609, USA.
3