A0530
Title: Local effects of continuous instruments without positivity
Authors: Luke Keele - University of Pennsylvania (United States) [presenting]
Abstract: Instrumental variables are a popular study design for the estimation of treatment effects in the presence of unobserved confounders. In the canonical instrumental variables design, the instrument is a binary variable. In many settings, however, the instrument is continuous. Standard estimation methods can be applied with continuous instruments, but they require strong assumptions. While recent work has introduced more flexible estimation approaches, these methods require a positivity assumption that is implausible in many applications. A novel family of causal estimands is derived using stochastic dynamic interventions that allow a range of intervention distributions that are continuous with respect to the observed distribution of the instrument. These estimands focus on a specific local effect but do not require a positivity assumption. Next, doubly robust estimators are developed for these estimands that allow for the estimation of the nuisance functions via nonparametric estimators. Empirical process theory and sample splitting are used to derive asymptotic properties of the proposed estimators under weak conditions. In addition, methods are derived for profiling the principal strata, as well as a method of sensitivity analysis. The methods are evaluated via simulation and demonstrate their feasibility using an application on the effectiveness of surgery for specific emergency conditions.
