Two-stage least squares with a randomly right censored outcome

TL;DR

This paper introduces a modified two-stage least squares method for estimating causal effects when the outcome variable is subject to random right censoring, using inverse probability weights to ensure consistency.

Contribution

It proposes a weighted 2SLS approach that accounts for censoring, providing theoretical guarantees and demonstrating good finite sample performance.

Findings

Estimator is consistent and asymptotically normal.

Performs well in finite samples based on simulations.

Addresses causal inference with censored outcomes.

Abstract

This note develops a simple two-stage least squares (2SLS) procedure to estimate the causal effect of some endogenous regressors on a randomly right censored outcome in the linear model. The proposal replaces the usual ordinary least squares regressions of the standard 2SLS by weighted least squares regressions. The weights correspond to the inverse probability of censoring. We show consistency and asymptotic normality of the estimator. The estimator exhibits good finite sample performances in simulations.