Piecewise Linear Instrumental Variable Estimation of Causal Influence

Richard Scheines, Gregory F. Cooper, Changwon Yoo, Tianjiao Chu
Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics, PMLR R3:265-271, 2001.

Abstract

Instrumental Variable (IV) estimation is a powerful strategy for estimating causal influence, even in the presence of confounding. Standard IV estimation requires that the relationships between variables is linear. Here we relax the linearity requirement by constructing a piecewise linear IV estimator. Simulation studies show that when the causal influence of $X$ on $Y$ is non-linear, the piecewise linear is an improvement.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR3-scheines01a, title = {Piecewise Linear Instrumental Variable Estimation of Causal Influence}, author = {Scheines, Richard and Cooper, Gregory F. and Yoo, Changwon and Chu, Tianjiao}, booktitle = {Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics}, pages = {265--271}, year = {2001}, editor = {Richardson, Thomas S. and Jaakkola, Tommi S.}, volume = {R3}, series = {Proceedings of Machine Learning Research}, month = {04--07 Jan}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/r3/scheines01a/scheines01a.pdf}, url = {https://proceedings.mlr.press/r3/scheines01a.html}, abstract = {Instrumental Variable (IV) estimation is a powerful strategy for estimating causal influence, even in the presence of confounding. Standard IV estimation requires that the relationships between variables is linear. Here we relax the linearity requirement by constructing a piecewise linear IV estimator. Simulation studies show that when the causal influence of $X$ on $Y$ is non-linear, the piecewise linear is an improvement.}, note = {Reissued by PMLR on 31 March 2021.} }
Endnote
%0 Conference Paper %T Piecewise Linear Instrumental Variable Estimation of Causal Influence %A Richard Scheines %A Gregory F. Cooper %A Changwon Yoo %A Tianjiao Chu %B Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2001 %E Thomas S. Richardson %E Tommi S. Jaakkola %F pmlr-vR3-scheines01a %I PMLR %P 265--271 %U https://proceedings.mlr.press/r3/scheines01a.html %V R3 %X Instrumental Variable (IV) estimation is a powerful strategy for estimating causal influence, even in the presence of confounding. Standard IV estimation requires that the relationships between variables is linear. Here we relax the linearity requirement by constructing a piecewise linear IV estimator. Simulation studies show that when the causal influence of $X$ on $Y$ is non-linear, the piecewise linear is an improvement. %Z Reissued by PMLR on 31 March 2021.
APA
Scheines, R., Cooper, G.F., Yoo, C. & Chu, T.. (2001). Piecewise Linear Instrumental Variable Estimation of Causal Influence. Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research R3:265-271 Available from https://proceedings.mlr.press/r3/scheines01a.html. Reissued by PMLR on 31 March 2021.

Related Material