eduzhai > Applied Sciences > Computer Science >

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

  • KanKan
  • (0) Download
  • 20210426
  • Save

... pages left unread,continue reading

Document pages: 31 pages

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

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...