Cause-Effect Inference by Comparing Regression Errors
Patrick Bloebaum, Dominik Janzing, Takashi Washio, Shohei Shimizu, Bernhard Schoelkopf
Abstract
We address the problem of inferring the causal relation between two variables by comparing the least-squares errors of the predictions in both possible causal directions. Under the assumption of an independence between the function relating cause and effect, the conditional noise distribution, and the distribution of the cause, we show that the errors are smaller in causal direction if both variables are equally scaled and the causal relation is close to deterministic. Based on this, we provide an easily applicable method that only requires a regression in both possible causal directions. The performance of this method is compared with different related causal inference methods in various artificial and real-world data sets.
BibTeX
@InProceedings{pmlr-v84-bloebaum18a,
title = {Cause-Effect Inference by Comparing Regression Errors},
author = {Bloebaum, Patrick and Janzing, Dominik and Washio, Takashi and Shimizu, Shohei and Schoelkopf, Bernhard},
booktitle = {Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics},
pages = {900--909},
year = {2018},
editor = {Storkey, Amos and Perez-Cruz, Fernando},
volume = {84},
series = {Proceedings of Machine Learning Research},
month = {09--11 Apr},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v84/bloebaum18a/bloebaum18a.pdf},
url = {https://proceedings.mlr.press/v84/bloebaum18a.html},
abstract = {We address the problem of inferring the causal relation between two variables by comparing the least-squares errors of the predictions in both possible causal directions. Under the assumption of an independence between the function relating cause and effect, the conditional noise distribution, and the distribution of the cause, we show that the errors are smaller in causal direction if both variables are equally scaled and the causal relation is close to deterministic. Based on this, we provide an easily applicable method that only requires a regression in both possible causal directions. The performance of this method is compared with different related causal inference methods in various artificial and real-world data sets.}
}