|
Activity Number:
|
298
- Model/Variable Selection and Model Evaluation
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 30, 2019 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biometrics Section
|
|
Abstract #306847
|
|
|
Title:
|
Variable Selection in Enriched Dirichlet Process with Applications to Causal Inference
|
|
Author(s):
|
Kumaresh Dhara* and Michael Daniels
|
|
Companies:
|
University of Florida and University of Florida
|
|
Keywords:
|
Causal Inference ;
Enriched Dirichlet Process;
Variable Selection
|
|
Abstract:
|
Dirichlet process mixtures are often used to model the joint distribution of a response and predictors. However, the clusters formed when fitting the model often depends heavily on the covariates. Enriched Dirichlet process priors (EDP) overcomes these issues by modeling the joint distribution of response and predictors using a nested structure. EDP has been recently used in causal inference. It is common that a large number of covariates are available for modeling the response but only a few of them are important. In this paper, we propose a variable selection approach while using an enriched Dirichlet process. Removing irrelevant covariates helps in efficient and simpler modeling of the joint structure of the response and covariates.
|
Authors who are presenting talks have a * after their name.
