Abstract
Researchers often need to determine whether a specific exposure, or something else, caused an individual’s outcome. To answer questions of causality in which the exposure and outcome have already been observed, researchers have suggested estimating the probability of causation (PC). However, authors disagree about the proper definitions and identification strategies for PC, and current estimation methods for PC make strong parametric assumptions, or are inefficient and do not easily yield inferential tools. I discuss the definition and identification assumptions of PC, derive a novel estimation method for PC under standard identifiability assumptions under which PC equals the excess risk ratio, and derive an influence-function-based nonparametric estimator for PC, which allows for simple interpretation and valid inference by making only weak structural assumptions. Finally, I present an application of my estimation method.
Cuellar, Maria, Causal Reasoning and Data Analysis in the Law: Definition, Estimation, and Usage of the Probability of Causation (May 31, 2018).
First posted 2018-06-01 08:33:46
Leave a Reply