Optimal policy learning using Stata

Abstract

In this article, I introduce the package opl for optimal policy learning, facilitating ex ante policy impact evaluation within the Stata environment. Despite theoretical progress, practical implementations of policy-learning algorithms are still poor within popular statistical software. To address this limitation, opl implements three popular policy-learning algorithms in Stata—threshold based, linear combination, and fixed-depth decision tree—and provides practical demonstrations of them using a real dataset. I also present policy-scenario development proposing a menu strategy, which is particularly useful when selection variables are affected by welfare monotonicity. Overall, this article contributes to bridging the gap between theoretical advancements and practical applications in the field of policy learning.

Keywords

st0774 opl make_cate opl_dt_c opl_dt opl_lc_c opl_lc opl_tb_c opl_tb ex ante policy evaluation optimal policy learning optimal treatment assignment machine learning conditional average treatment effect

Get full access to this article

View all access options for this article.

References

Athey

Wager

. 2021. Policy learning with observational data. Econometrica 89: 133–161. 10.3982/ECTA15732.

Bhattacharya

Dupas

. 2012. Inferring welfare maximizing treatment assignment under budget constraints. Journal of Econometrics 167: 168–196. https://doi.org/10.1016/j.jeconom.2011.11.007 .

Cerulli

2022a. Econometric Evaluation of Socio-Economic Programs: Theory and Applications. Vol. 49 of Advanced Studies in Theoretical and Applied Econometrics. Berlin: Springer. 10.1007/978-3-662-65945-8.

——. 2022b. r_ml_stata_cv: Stata module to implement machine learning regression in Stata. Statistical Software Components S459054, Department of Economics, Boston College. https://ideas.repec.org/c/boc/bocode/s459054.html .

——. 2023a. Fundamentals of Supervised Machine Learning: With Applications in Python, R, and Stata. Statistics and Computing. Cham, Switzerland: Springer. https://doi.org/10.1007/978-3-031-41337-7.

——. 2023b. Optimal treatment assignment of a threshold-based policy: Empirical protocol and related issues. Applied Economics Letters 30: 1010–1017. https://doi.org/10.1080/13504851.2022.2032577 .

Chernozhukov

Chetverikov

Demirer

Duflo

Hansen

Newey

Robins

. 2018. Double/debiased machine learning for treatment and structural parameters. Econometrics Journal 21: C1–C68. 10.1111/ectj.12097.

Dehejia

R. H

. 2005. Program evaluation as a decision problem. Journal ofEconometrics 125: 141–173. 10.1016/j.jeconom.2004.04.006.

Dehejia

R. H.

Wahba

. 1999. Causal effects in nonexperimental studies: Reevaluating the evaluation of training programs. Journal of the American Statistical Association 94: 1053–1062. 10.1080/01621459.1999.10473858.

10.

Garbero

Sakos

Cerulli

. 2023. Towards data-driven project design: Providing optimal treatment rules for development projects. Socio-Economic Planning Sciences 89: art. 101618. 10.1016/j.seps.2023.101618.

11.

Hastie

Tibshirani

Friedman

. 2009. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer Series in Statistics, 2nd ed. New York: Springer. 10.1007/978-0-387-84858-7.

12.

Kennedy

E. H

. 2023. Towards optimal doubly robust estimation of heterogeneous causal effects. Electronic Journal of Statistics 17: 3008–3049. 10.1214/23-EJS2157.

13.

Kitagawa

Tetenov

. 2018. Who should be treated? Empirical welfare maximization methods for treatment choice. Econometrica 86: 591–616. https://doi.org/10.3982/ECTA13288 .

14.

Künzel

S. R.

Sekhon

J. S.

Bickel

P. J.

. 2019. Metalearners for estimating heterogeneous treatment effects using machine learning. Proceedings of the National Academy of Sciences 116: 4156–4165. 10.1073/pnas.1804597116.

15.

LaLonde

R. J

. 1986. Evaluating the econometric evaluations of training programs with experimental data. American Economic Review 76: 604–620.

16.

Manski

C. F

. 2004. Statistical treatment rules for heterogeneous populations. Econometrica 72: 1221–1246. 10.1111/j.1468-0262.2004.00530.x.

17.

May

P. J

. 1992. Policy learning and failure. Journal of Public Policy 12: 331–354. https://doi.org/10.1017/S0143814X00005602 .

18.

Mbakop

Tabord-Meehan

. 2016. Model selection for treatment choice: Penalized welfare maximization. arXiv:1609.03167v6 [math.ST], 10.48550/arXiv.1609.03167.

19.

Nie

Brunskill

Wager

. 2019. Learning when-to-treat policies. arXiv:1905.09751 [stat.ME], 10.48550/arXiv.1905.09751.

20.

Semenova

Chernozhukov

. 2021. Debiased machine learning of conditional average treatment effects and other causal functions. Econometrics Journal 24: 264–289. 10.1093/ectj/utaa027.

21.

Sutherland

W. J.

Burgman

. 2015. Policy advice: Use experts wisely. Nature 526: 317–318. 10.1038/526317a.

22.

Zhou

Athey

Wager

. 2023. Offline multi-action policy learning: Generalization and optimization. Operations Research 71: 148–183. 10.1287/opre.2022.2271.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB

0.00 MB

0.04 MB

0.00 MB