Subjective Causality in Choice

TL;DR

This paper models how agents use subjective causal beliefs, represented by DAGs, to make choices based on observational data, revealing the causal reasoning behind their decisions and how their behavior influences their inferences.

Contribution

It introduces a method to identify agents' subjective causal models from their choice behavior and observational data, linking causal reasoning to decision-making processes.

Findings

Agents' choice rules reveal their causal beliefs and confounding adjustments.

The model provides necessary and sufficient conditions to test compatibility of behavior with subjective causal models.

Behavior influences inferences, creating a feedback loop in causal reasoning.

Abstract

Choices based on observational data depend on beliefs about which correlations reflect causality. An agent predicts the consequence of available actions using a dataset and her subjective beliefs about causality represented by a directed acyclic graph (DAG). We identify her DAG from her random choice rule. Her choices reveal the chains of causal reasoning that she undertakes and the confounding variables she adjusts for, and these pin down her model. When her choices generate the dataset used, her behavior affects her inferences, which in turn affect her choices. We provide necessary and sufficient conditions for testing whether her behavior is compatible with such a model.