Probabilities of Causation with Nonbinary Treatment and Effect
This work addresses a methodological gap in causal inference for nonbinary scenarios, offering incremental theoretical extensions with potential applications in decision-making.
The paper tackles the problem of estimating probabilities of causation for multivalued treatments and effects, extending Tian and Pearl's binary bounds to provide theoretical bounds for all types of causation probabilities, with simulation studies evaluating their informativeness.
This paper deals with the problem of estimating the probabilities of causation when treatment and effect are not binary. Tian and Pearl derived sharp bounds for the probability of necessity and sufficiency (PNS), the probability of sufficiency (PS), and the probability of necessity (PN) using experimental and observational data. In this paper, we provide theoretical bounds for all types of probabilities of causation to multivalued treatments and effects. We further discuss examples where our bounds guide practical decisions and use simulation studies to evaluate how informative the bounds are for various combinations of data.