Distinguishing Cause from Effect Using Quantiles: Bivariate Quantile Causal Discovery

Natasa Tagasovska, Valérie Chavez-Demoulin, Thibault Vatter
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:9311-9323, 2020.

Abstract

Causal inference using observational data is challenging, especially in the bivariate case. Through the minimum description length principle, we link the postulate of independence between the generating mechanisms of the cause and of the effect given the cause to quantile regression. Based on this theory, we develop Bivariate Quantile Causal Discovery (bQCD), a new method to distinguish cause from effect assuming no confounding, selection bias or feedback. Because it uses multiple quantile levels instead of the conditional mean only, bQCD is adaptive not only to additive, but also to multiplicative or even location-scale generating mechanisms. To illustrate the effectiveness of our approach, we perform an extensive empirical comparison on both synthetic and real datasets. This study shows that bQCD is robust across different implementations of the method (i.e., the quantile regression), computationally efficient, and compares favorably to state-of-the-art methods.

Cite this Paper


BibTeX
@InProceedings{pmlr-v119-tagasovska20a, title = {Distinguishing Cause from Effect Using Quantiles: Bivariate Quantile Causal Discovery}, author = {Tagasovska, Natasa and Chavez-Demoulin, Val{\'e}rie and Vatter, Thibault}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {9311--9323}, year = {2020}, editor = {III, Hal Daumé and Singh, Aarti}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/tagasovska20a/tagasovska20a.pdf}, url = {https://proceedings.mlr.press/v119/tagasovska20a.html}, abstract = {Causal inference using observational data is challenging, especially in the bivariate case. Through the minimum description length principle, we link the postulate of independence between the generating mechanisms of the cause and of the effect given the cause to quantile regression. Based on this theory, we develop Bivariate Quantile Causal Discovery (bQCD), a new method to distinguish cause from effect assuming no confounding, selection bias or feedback. Because it uses multiple quantile levels instead of the conditional mean only, bQCD is adaptive not only to additive, but also to multiplicative or even location-scale generating mechanisms. To illustrate the effectiveness of our approach, we perform an extensive empirical comparison on both synthetic and real datasets. This study shows that bQCD is robust across different implementations of the method (i.e., the quantile regression), computationally efficient, and compares favorably to state-of-the-art methods.} }
Endnote
%0 Conference Paper %T Distinguishing Cause from Effect Using Quantiles: Bivariate Quantile Causal Discovery %A Natasa Tagasovska %A Valérie Chavez-Demoulin %A Thibault Vatter %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-tagasovska20a %I PMLR %P 9311--9323 %U https://proceedings.mlr.press/v119/tagasovska20a.html %V 119 %X Causal inference using observational data is challenging, especially in the bivariate case. Through the minimum description length principle, we link the postulate of independence between the generating mechanisms of the cause and of the effect given the cause to quantile regression. Based on this theory, we develop Bivariate Quantile Causal Discovery (bQCD), a new method to distinguish cause from effect assuming no confounding, selection bias or feedback. Because it uses multiple quantile levels instead of the conditional mean only, bQCD is adaptive not only to additive, but also to multiplicative or even location-scale generating mechanisms. To illustrate the effectiveness of our approach, we perform an extensive empirical comparison on both synthetic and real datasets. This study shows that bQCD is robust across different implementations of the method (i.e., the quantile regression), computationally efficient, and compares favorably to state-of-the-art methods.
APA
Tagasovska, N., Chavez-Demoulin, V. & Vatter, T.. (2020). Distinguishing Cause from Effect Using Quantiles: Bivariate Quantile Causal Discovery. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:9311-9323 Available from https://proceedings.mlr.press/v119/tagasovska20a.html.

Related Material