Distinguishing Cause from Effect Based on Exogeneity

By Kun Zhang, Jiji Zhang, Bernhard Schölkopf. 2015

36 citations recorded by the source corpus.

Recent developments in structural equation modeling have produced several methods that can usually distinguish cause from effect in the two-variable case. For that purpose, however, one has to impose substantial structural constraints or smoothness assumptions on the functional causal models. In this paper, we consider the problem of determining the causal direction from a related but different point of view, and propose a new framework for causal direction determination. We show that it is possible to perform causal inference based on the condition that the cause is "exogenous" for the parameters involved in the generating process from the cause to the effect. In this way, we avoid the structural constraints required by the SEM-based approaches. In particular, we exploit nonparametric methods to estimate marginal and conditional distributions, and propose a bootstrap-based approach to test for the exogeneity condition; the testing results indicate the causal direction between two variables. The proposed method is validated on both synthetic and real data.

Subfields: Artificial Intelligence, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Management Science and Operations Research

Citations in the corpus

Learning Independent Causal Mechanisms (2017, methodology): “The IM assumption discussed in the introduction can be used for identiﬁcation of causal models (Daniuˇsis et al., 2010; Zhang et al., 2015), but causality has also proven a useful tool for discussing and understanding machine learning in the non-i.i.d. regime.”
Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows (2017, neutral): “Problem 2, as a sub-problem of causal discovery, exploits a generalized notion of the invariance property [10] or the exogeneity property [2], [13] of causal systems: if there is no nonstationary confounder for Vi and Vj , then the causal mechanisms, represented by the conditional distributions P (V”
Error asymmetry in causal and anticausal regression (2016, background): “This assumption is also related to the autonomous data generation process and exogeneity (Pearl 2009; Zhang et al. 2015).”
Causal Transfer in Machine Learning (2015, background): “This notion is closely related to exogeneity [Zhang et al., 2015b], which roughly states that a causal mechanism mapping a cause X to Y should not depend on the distribution of X .”
Causal Discovery and Forecasting in Nonstationary Environments with State-Space Models (2019, background): “…postulated that if there is no confounder, the marginal distribution P (cause) and the conditional distribution P (effect|cause) represent independent mechanisms of nature (Pearl, 2000; Janzing & Schölkopf, 2010), which is related to the exogeneity notion (Engle et al., 1983; Zhang et al., 2015b).”
Multi-domain Causal Structure Learning in Linear Systems (2018, background): “This can be viewed as a realization of the modularity property of causal systems [Pea09], and as the dynamic counterpart of the principle of independent mechanisms, which states that causal modules are algorithmically independent [DJM10], or the exogeneity property of the causal system [ZZS15].”
Invariant Models for Causal Transfer Learning (2015, background): “Moreover, Zhang et al. (2015a) argue that the availability of multiple domains is suﬃcient to drop this previous assumption when the distribution of Y k and the conditional X k | Y k change independently.”