Invariant Causal Prediction for Sequential Data

ABSTRACT

We investigate the problem of inferring the causal predictors of a response Y from a set of d explanatory variables (X¹, …, X^d). Classical ordinary least-square regression includes all predictors that reduce the variance of Y. Using only the causal predictors instead leads to models that have the advantage of remaining invariant under interventions; loosely speaking they lead to invariance across different “environments” or “heterogeneity patterns.” More precisely, the conditional distribution of Y given its causal predictors is the same for all observations, provided that there are no interventions on Y. Recent work exploits such a stability to infer causal relations from data with different but known environments. We show that even without having knowledge of the environments or heterogeneity pattern, inferring causal relations is possible for time-ordered (or any other type of sequentially ordered) data. In particular, this allows detecting instantaneous causal relations in multivariate linear time series, which is usually not the case for Granger causality. Besides novel methodology, we provide statistical confidence bounds and asymptotic detection results for inferring causal predictors, and present an application to monetary policy in macroeconomics. Supplementary materials for this article are available online.

Keywords:

• Causal structure learning
• Change point model
• Chow statistic
• Instantaneous causal effects
• Monetary policy

Acknowledgments

The authors thank Nicolai Meinshausen and anonymous reviewers for helpful discussions and constructive comments.

Additional information

Funding

Niklas Pfister is partially funded by Swiss National Science Foundation (SNSF) no. 200021_153504 and Jonas Peters is partially funded by a research grant (18968) from VILLUM FONDEN.