RNN-based counterfactual prediction, with an application to homestead policy and public schooling

TL;DR

This paper introduces an RNN-based approach for counterfactual prediction in policy analysis, leveraging temporal dependencies to estimate long-term impacts, demonstrated through a case study on U.S. homestead policy and school spending.

Contribution

It develops a novel RNN framework for counterfactual estimation that captures nonlinear and negative interactions in panel data, improving policy impact assessments.

Findings

RNNs effectively model temporal dependencies in panel data.

The method accurately predicts counterfactual outcomes for policy evaluation.

Application to homestead policy reveals significant long-term effects on school spending.

Abstract

This paper proposes a method for estimating the effect of a policy intervention on an outcome over time. We train recurrent neural networks (RNNs) on the history of control unit outcomes to learn a useful representation for predicting future outcomes. The learned representation of control units is then applied to the treated units for predicting counterfactual outcomes. RNNs are specifically structured to exploit temporal dependencies in panel data, and are able to learn negative and nonlinear interactions between control unit outcomes. We apply the method to the problem of estimating the long-run impact of U.S. homestead policy on public school spending.