|
Activity Number:
|
169
- Advanced Bayesian Topics (Part 2)
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 10, 2021 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
|
Abstract #318028
|
|
|
Title:
|
Bayesian Propensity Score Analysis for Misclassified Multinomial Data
|
|
Author(s):
|
Yuhan Ma* and Joon Jin Song
|
|
Companies:
|
Baylor University and Baylor University
|
|
Keywords:
|
Causal Inference;
Misclassification;
Propensity Score Analysis;
Multinomial
|
|
Abstract:
|
A Bayesian propensity score regression model has been proposed to estimate the causal effect on the observational data with the misclassified multinominal response. This model estimates the propensity scores for balancing the observed variables and corrects for misclassification in the responses. To eliminate the non-identifiability issue, the informative Dirichlet prior has been specified on the parameters of misclassification probabilities. The proposed model is compared with the model without correcting misclassification in the simulation study. It was found that the performance of the proposed model has been improved with smaller bias and the coverage probabilities closer to the given nominal value. We confirmed that ignoring the misclassification in the response leads to biased estimation of the treatment effect. We apply the proposed model for the study of estimating the religious impact on the choice of contraception method.
|
Authors who are presenting talks have a * after their name.
