Counterfactual Analysis in Dynamic Latent State Models

TL;DR

This paper introduces an optimization-based framework for counterfactual analysis in dynamic models with hidden states, providing bounds on counterfactuals despite unknown causal mechanisms, demonstrated through a breast cancer case study.

Contribution

It is the first to compute bounds on counterfactuals in dynamic latent-state models, integrating causality, state-space models, and optimization.

Findings

First to compute bounds on counterfactuals in dynamic latent models

Applied framework to breast cancer case study

Provides a method to handle unknown causal mechanisms

Abstract

We provide an optimization-based framework to perform counterfactual analysis in a dynamic model with hidden states. Our framework is grounded in the ``abduction, action, and prediction'' approach to answer counterfactual queries and handles two key challenges where (1) the states are hidden and (2) the model is dynamic. Recognizing the lack of knowledge on the underlying causal mechanism and the possibility of infinitely many such mechanisms, we optimize over this space and compute upper and lower bounds on the counterfactual quantity of interest. Our work brings together ideas from causality, state-space models, simulation, and optimization, and we apply it on a breast cancer case study. To the best of our knowledge, we are the first to compute lower and upper bounds on a counterfactual query in a dynamic latent-state model.