Semiparametric Estimation of Long-Term Treatment Effects

TL;DR

This paper introduces semiparametric methods to estimate long-term treatment effects by combining experimental short-term outcomes with observational data, addressing delays in observing long-term results.

Contribution

It develops new semiparametric estimators and derives efficiency bounds for long-term effect estimation, improving upon existing methods.

Findings

Estimators perform well in finite samples based on simulation results.

The methods effectively integrate experimental and observational data.

Efficiency bounds guide optimal estimator construction.

Abstract

Long-term outcomes of experimental evaluations are necessarily observed after long delays. We develop semiparametric methods for combining the short-term outcomes of experiments with observational measurements of short-term and long-term outcomes, in order to estimate long-term treatment effects. We characterize semiparametric efficiency bounds for various instances of this problem. These calculations facilitate the construction of several estimators. We analyze the finite-sample performance of these estimators with a simulation calibrated to data from an evaluation of the long-term effects of a poverty alleviation program.