A1014
Title: Distributional instrumental variable method
Authors: Xinwei Shen - University of Washington (United States) [presenting]
Abstract: The instrumental variable (IV) approach is commonly used to infer causal effects in the presence of unmeasured confounding. Existing methods typically aim to estimate the mean causal effects, whereas a few other methods focus on quantile treatment effects. The aim is to estimate the entire interventional distribution, which yields the classical causal estimands as functionals. A method called distributional instrumental variable (DIV) is proposed, which uses generative modelling in a nonlinear IV setting. Identifiability of the interventional distribution is established under general assumptions, and an `under-identified' case is demonstrated, where DIV can identify the causal effects while two-step least squares fails to. The empirical results show that the DIV method performs well for a broad range of simulated data, exhibiting advantages over existing IV approaches in terms of the identifiability and estimation error of the mean or quantile treatment effects. Furthermore, DIV is applied to an economic data set to examine the causal relation between institutional quality and economic development, and the results align well with the original study. DIV is also applied to a single-cell data set, where we study the generalizability and stability in predicting gene expression under unseen interventions.
