Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A comparative study of design-based and analysis-based approaches to causal inference with observational data

A comparative study of design-based and analysis-based approaches to causal inference with... Causal inference with observational data is a central goal in many fields. Propensity score methods are design-based approaches that try to ensure covariate balance without using information from the outcome variables. Analysis-based approaches, such as the Bayesian Additive Regression Tree and the Causal Forest, bypass the issue of covariate balance, and directly model the outcomes. We use a Monte Carlo simulation to study the performance of these two types of approaches. Some of the simulation scenarios involve large number of covariates relative to the number of observations. We find that the analysis-based approaches can yield very poor performance, without any warning about not enough overlap between the covariate distributions for the treated and control groups. In contrast, the propensity score methods provide warning about not enough overlap, but such warning could be overly-cautious when there is enough overlap. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Biostatistics & Epidemiology Taylor & Francis

A comparative study of design-based and analysis-based approaches to causal inference with observational data

Zhang, Junni L.
Biostatistics & Epidemiology , Volume OnlineFirst: 10 – Oct 28, 2021
Download PDF

A comparative study of design-based and analysis-based approaches to causal inference with observational data

Abstract

Causal inference with observational data is a central goal in many fields. Propensity score methods are design-based approaches that try to ensure covariate balance without using information from the outcome variables. Analysis-based approaches, such as the Bayesian Additive Regression Tree and the Causal Forest, bypass the issue of covariate balance, and directly model the outcomes. We use a Monte Carlo simulation to study the performance of these two types of approaches. Some of the...
Loading next page...
 
/lp/taylor-francis/a-comparative-study-of-design-based-and-analysis-based-approaches-to-ZuHJn0Pbit
Publisher
Taylor & Francis
Copyright
© 2021 International Biometric Society – Chinese Region
ISSN
2470-9379
eISSN
2470-9360
DOI
10.1080/24709360.2021.1992246
Publisher site
See Article on Publisher Site

Abstract

Causal inference with observational data is a central goal in many fields. Propensity score methods are design-based approaches that try to ensure covariate balance without using information from the outcome variables. Analysis-based approaches, such as the Bayesian Additive Regression Tree and the Causal Forest, bypass the issue of covariate balance, and directly model the outcomes. We use a Monte Carlo simulation to study the performance of these two types of approaches. Some of the simulation scenarios involve large number of covariates relative to the number of observations. We find that the analysis-based approaches can yield very poor performance, without any warning about not enough overlap between the covariate distributions for the treated and control groups. In contrast, the propensity score methods provide warning about not enough overlap, but such warning could be overly-cautious when there is enough overlap.

Journal

Biostatistics & Epidemiology – Taylor & Francis

Published: Oct 28, 2021

Keywords: Average treatment effect on the treated; conditional average treatment effect on the treated; propensity score matching; lasso; causal tree

References

  • The central role of the propensity score in observational studies for causal effects
  • Using propensity scores to help design observational studies: application to the tobacco litigation
  • Bayesian nonparametric modeling for causal inference
  • Estimation and inference of heterogeneous treatment effects using random forests
  • Discussion of ‘randomization analysis of experimental data: the Fisher randomization test’ by D. Basu
  • Constructing a control group using multivariate matched sampling methods that incorporate the propensity score
  • Large sample properties of matching estimators for average treatment effects
  • Regression shrinkage and selection via the lasso
  • -regularization path algorithm for generalised linear models
  • Comparing the performance of propensity score methods in healthcare database studies with rare outcomes
  • Does the estimation of the propensity score by machine learning improve matching estimation? The case of Germany's programmes for long term unemployed
  • Bart: Bayesian additive regression trees