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 predic-
tion” 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 under-
lying causal mechanism and the possibility of in-
finitely 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.