Causal Inference with the Instrumental Variable Approach and Bayesian Nonparametric Machine Learning
Machine Learning
2021-02-03 v1 Machine Learning
Abstract
We provide a new flexible framework for inference with the instrumental variable model. Rather than using linear specifications, functions characterizing the effects of instruments and other explanatory variables are estimated using machine learning via Bayesian Additive Regression Trees (BART). Error terms and their distribution are inferred using Dirichlet Process mixtures. Simulated and real examples show that when the true functions are linear, little is lost. But when nonlinearities are present, dramatic improvements are obtained with virtually no manual tuning.
Cite
@article{arxiv.2102.01199,
title = {Causal Inference with the Instrumental Variable Approach and Bayesian Nonparametric Machine Learning},
author = {Robert E. McCulloch and Rodney A. Sparapani and Brent R. Logan and Purushottam W. Laud},
journal= {arXiv preprint arXiv:2102.01199},
year = {2021}
}
Comments
33 pages, 7 figures