Inference on Local Average Treatment Effects for Misclassified Treatment

TL;DR

This paper develops a method to accurately estimate local average treatment effects when the binary treatment variable is misclassified due to measurement error, using instrumental variables and GMM estimation.

Contribution

It introduces a novel identification strategy for treatment effects under measurement error and provides a GMM-based estimation approach with valid inference.

Findings

The proposed method corrects bias caused by measurement error.

Simulation studies show improved estimation accuracy.

Empirical application demonstrates practical usefulness.

Abstract

We develop point-identification for the local average treatment effect when the binary treatment contains a measurement error. The standard instrumental variable estimator is inconsistent for the parameter since the measurement error is non-classical by construction. We correct the problem by identifying the distribution of the measurement error based on the use of an exogenous variable that can even be a binary covariate. The moment conditions derived from the identification lead to generalized method of moments estimation with asymptotically valid inferences. Monte Carlo simulations and an empirical illustration demonstrate the usefulness of the proposed procedure.