Online Program Home
My Program

Abstract Details

Activity Number: 472 - Statistical Methods for Causal Inference
Type: Contributed
Date/Time: Wednesday, July 31, 2019 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistics in Epidemiology
Abstract #304645 Presentation 1 Presentation 2
Title: Detecting Heterogeneous Treatment Effect with Instrumental Variables in Causal Inference
Author(s): Michael Johnson* and Hyunseung Kang
Companies: University of Wisconsin-Madison and University of Wisconsin-Madison
Keywords: Causal Inference; Heterogeneous Treatment Effect; Instrumental Variables; Machine Learning; Matching

There is an increasing interest in methods estimating heterogeneity in causal effects in randomized and observational studies. However, little research has been conducted to understand heterogeneity in an instrumental variables study. In this work, we present a method to estimate heterogeneous causal effects using an instrumental variable approach. The method has two parts. The first part uses subject-matter knowledge and interpretable machine learning techniques, such as classification and regression trees and penalized regression, to partition the data set into potential subgroups with heterogeneous treatment effects. The second part tests for heterogeneous treatment effect within these partitions. To control for the concerns of using one data set to find partitions and make statistical inference, sample-splitting and a recent method of taking the absolute value of the outcome with Bonferroni correction and closed testing are applied to strongly control for familywise error rate. We conducted this method on a real data set example on the effect of malaria on stunted child growth, which showed evidence of heterogeneity in children with and without mosquito nets for protection.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program