Foundations of Structural Causal Models with Cycles and Latent Variables

TL;DR

This paper extends the theory of structural causal models to include cycles and latent variables, identifying conditions under which key properties hold, and introduces simple SCMs that generalize acyclic models.

Contribution

It generalizes properties of SCMs to cyclic and latent variable settings, introducing simple SCMs that retain many desirable features of acyclic models.

Findings

Many properties of acyclic SCMs do not hold in cyclic cases without specific conditions.

Simple SCMs extend acyclic models to cycles while preserving key properties.

The paper lays the theoretical groundwork for causal modeling with complex SCMs.

Abstract

Structural causal models (SCMs), also known as (nonparametric) structural equation models (SEMs), are widely used for causal modeling purposes. In particular, acyclic SCMs, also known as recursive SEMs, form a well-studied subclass of SCMs that generalize causal Bayesian networks to allow for latent confounders. In this paper, we investigate SCMs in a more general setting, allowing for the presence of both latent confounders and cycles. We show that in the presence of cycles, many of the convenient properties of acyclic SCMs do not hold in general: they do not always have a solution; they do not always induce unique observational, interventional and counterfactual distributions; a marginalization does not always exist, and if it exists the marginal model does not always respect the latent projection; they do not always satisfy a Markov property; and their graphs are not always consistent with their causal semantics. We prove that for SCMs in general each of these properties does hold under certain solvability conditions. Our work generalizes results for SCMs with cycles that were only known for certain special cases so far. We introduce the class of simple SCMs that extends the class of acyclic SCMs to the cyclic setting, while preserving many of the convenient properties of acyclic SCMs. With this paper we aim to provide the foundations for a general theory of statistical causal modeling with SCMs.