Causal Inference with the Instrumental Variable Approach and Bayesian Nonparametric Machine Learning

Causal Inference with the Instrumental Variable Approach and Bayesian Nonparametric Machine Learning. McCulloch, R. E., Sparapani, R. A., Logan, B. R., & Laud, P. W. February, 2021. arXiv:2102.01199 [cs, stat]

Paper abstract bibtex

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.

@misc{mcculloch_causal_2021,
	title = {Causal {Inference} with the {Instrumental} {Variable} {Approach} and {Bayesian} {Nonparametric} {Machine} {Learning}},
	url = {http://arxiv.org/abs/2102.01199},
	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.},
	urldate = {2023-10-24},
	publisher = {arXiv},
	author = {McCulloch, Robert E. and Sparapani, Rodney A. and Logan, Brent R. and Laud, Purushottam W.},
	month = feb,
	year = {2021},
	note = {arXiv:2102.01199 [cs, stat]},
	keywords = {Computer Science - Machine Learning, Statistics - Machine Learning, 62P20 (Primary) 62G99 (Secondary)},
	annote = {Comment: 33 pages, 7 figures},
	file = {arXiv.org Snapshot:/Users/soumikp/Zotero/storage/CID6I4NL/2102.html:text/html;Full Text PDF:/Users/soumikp/Zotero/storage/FAK553JE/McCulloch et al. - 2021 - Causal Inference with the Instrumental Variable Ap.pdf:application/pdf},
}

Downloads: 0

{"_id":"9uD5NRXMj6wjtxeDN","bibbaseid":"mcculloch-sparapani-logan-laud-causalinferencewiththeinstrumentalvariableapproachandbayesiannonparametricmachinelearning-2021","author_short":["McCulloch, R. E.","Sparapani, R. A.","Logan, B. R.","Laud, P. W."],"bibdata":{"bibtype":"misc","type":"misc","title":"Causal Inference with the Instrumental Variable Approach and Bayesian Nonparametric Machine Learning","url":"http://arxiv.org/abs/2102.01199","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.","urldate":"2023-10-24","publisher":"arXiv","author":[{"propositions":[],"lastnames":["McCulloch"],"firstnames":["Robert","E."],"suffixes":[]},{"propositions":[],"lastnames":["Sparapani"],"firstnames":["Rodney","A."],"suffixes":[]},{"propositions":[],"lastnames":["Logan"],"firstnames":["Brent","R."],"suffixes":[]},{"propositions":[],"lastnames":["Laud"],"firstnames":["Purushottam","W."],"suffixes":[]}],"month":"February","year":"2021","note":"arXiv:2102.01199 [cs, stat]","keywords":"Computer Science - Machine Learning, Statistics - Machine Learning, 62P20 (Primary) 62G99 (Secondary)","annote":"Comment: 33 pages, 7 figures","file":"arXiv.org Snapshot:/Users/soumikp/Zotero/storage/CID6I4NL/2102.html:text/html;Full Text PDF:/Users/soumikp/Zotero/storage/FAK553JE/McCulloch et al. - 2021 - Causal Inference with the Instrumental Variable Ap.pdf:application/pdf","bibtex":"@misc{mcculloch_causal_2021,\n\ttitle = {Causal {Inference} with the {Instrumental} {Variable} {Approach} and {Bayesian} {Nonparametric} {Machine} {Learning}},\n\turl = {http://arxiv.org/abs/2102.01199},\n\tabstract = {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.},\n\turldate = {2023-10-24},\n\tpublisher = {arXiv},\n\tauthor = {McCulloch, Robert E. and Sparapani, Rodney A. and Logan, Brent R. and Laud, Purushottam W.},\n\tmonth = feb,\n\tyear = {2021},\n\tnote = {arXiv:2102.01199 [cs, stat]},\n\tkeywords = {Computer Science - Machine Learning, Statistics - Machine Learning, 62P20 (Primary) 62G99 (Secondary)},\n\tannote = {Comment: 33 pages, 7 figures},\n\tfile = {arXiv.org Snapshot:/Users/soumikp/Zotero/storage/CID6I4NL/2102.html:text/html;Full Text PDF:/Users/soumikp/Zotero/storage/FAK553JE/McCulloch et al. - 2021 - Causal Inference with the Instrumental Variable Ap.pdf:application/pdf},\n}\n\n","author_short":["McCulloch, R. E.","Sparapani, R. A.","Logan, B. R.","Laud, P. W."],"key":"mcculloch_causal_2021","id":"mcculloch_causal_2021","bibbaseid":"mcculloch-sparapani-logan-laud-causalinferencewiththeinstrumentalvariableapproachandbayesiannonparametricmachinelearning-2021","role":"author","urls":{"Paper":"http://arxiv.org/abs/2102.01199"},"keyword":["Computer Science - Machine Learning","Statistics - Machine Learning","62P20 (Primary) 62G99 (Secondary)"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"misc","biburl":"https://bibbase.org/f/Ceciz2iNjTZgQNtDc/mypubs_mar_2024.bib","dataSources":["m8Y57GfgnRrMKZTQS","epk5yKhDyD37NAsSC"],"keywords":["computer science - machine learning","statistics - machine learning","62p20 (primary) 62g99 (secondary)"],"search_terms":["causal","inference","instrumental","variable","approach","bayesian","nonparametric","machine","learning","mcculloch","sparapani","logan","laud"],"title":"Causal Inference with the Instrumental Variable Approach and Bayesian Nonparametric Machine Learning","year":2021}