Bayesian Analysis of Multiply Imputed Synthetic Data Under the Multiple Linear Regression Model

Thumbnail Image

Files

RRS2022-02.pdf (551.23 KB)

Links to Files

https://www.census.gov/content/dam/Census/library/working-papers/2022/adrm/RRS2022-02.pdf

Permanent Link

http://hdl.handle.net/11603/24774

Collections

UMBC Mathematics and Statistics Department
UMBC Faculty Collection

Author/Creator

Author/Creator ORCID

https://orcid.org/0000-0001-6361-8295

Date

2022-04-04

Type of Work

reports
Text

Department

Program

Citation of Original Publication

Guin, Abhishek.Bayesian Analysis of Multiply Imputed Synthetic Data Under the Multiple Linear Regression Model. Washington, D.C: U.S. Census Bureau, Apr. 4, 2022. https://www.census.gov/content/dam/Census/library/working-papers/2022/adrm/RRS2022-02.pdf

Rights

This is a work of the United States Government. In accordance with 17 U.S.C. 105, no copyright protection is available for such works under U.S. Law.
Public Domain Mark 1.0

Subjects

Abstract

In this paper we consider Bayesian inference of model parameters in a multiple linear regression model when the response variable is sensitive and the covariates are not, analysis being carried out based on multiple synthetic versions of the response variable. Two scenarios of synthetic data generation are considered - plug-in sampling method and posterior predictive sampling method. We also consider the case when part of the response is sensitive and describe how to carry out full Bayesian analysis based on multiply imputed data.