Abstract:Social science theories often postulate causal relationships among a set of variables or events. Although directed acyclic graphs (DAGs) are increasingly used to represent these theories, their full potential has not yet been realized in practice. As non-parametric causal models, DAGs require no assumptions about the functional form of the hypothesized relationships. Nevertheless, to simplify the task of empirical evaluation, researchers tend to invoke such assumptions anyway, even though they are typically arbitrary and do not reflect any theoretical content or prior knowledge. Moreover, functional form assumptions can engender bias, whenever they fail to accurately capture the complexity of the causal system under investigation. In this article, we introduce causal-graphical normalizing flows (cGNFs), a novel approach to causal inference that leverages deep neural networks to empirically evaluate theories represented as DAGs. Unlike conventional approaches, cGNFs model the full joint distribution of the data according to a DAG supplied by the analyst, without relying on stringent assumptions about functional form. In this way, the method allows for flexible, semi-parametric estimation of any causal estimand that can be identified from the DAG, including total effects, conditional effects, direct and indirect effects, and path-specific effects. We illustrate the method with a reanalysis of Blau and Duncan's (1967) model of status attainment and Zhou's (2019) model of conditional versus controlled mobility. To facilitate adoption, we provide open-source software together with a series of online tutorials for implementing cGNFs. The article concludes with a discussion of current limitations and directions for future development.

Invited Commentary: Where Do the Causal DAGs Come From?

Deep Learning With DAGs

DAG With Omitted Objects Displayed (DAGWOOD): A framework for revealing causal assumptions in DAGs

Credible causal inference beyond toy models

How to develop causal directed acyclic graphs for observational health research: a scoping review

Measurement error and information bias in causal diagrams: mapping epidemiological concepts and graphical structures

A review of some recent advances in causal inference

Supervised Whole DAG Causal Discovery

A biologist's guide to model selection and causal inference

Tutorial on directed acyclic graphs

Reinterpreting causal discovery as the task of predicting unobserved joint statistics

Depicting deterministic variables within directed acyclic graphs (DAGs): An aid for identifying and interpreting causal effects involving derived variables and compositional data

Causal DAG extraction from a library of books or videos/movies

Causal Data Integration

Causal Structure Learning: a Combinatorial Perspective

What Is a Causal Graph?

Personalized Binomial DAGs Learning with Network Structured Covariates

Depicting deterministic variables within directed acyclic graphs (DAGs): An aid for identifying and interpreting causal effects involving tautological associations, compositional data, and composite variables

On the Representation of Causal Background Knowledge and its Applications in Causal Inference

Argumentative Causal Discovery

Drawing and Analyzing Causal DAGs with DAGitty