Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
139 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Regret Analysis in Threshold Policy Design (2404.11767v2)

Published 17 Apr 2024 in econ.EM

Abstract: Threshold policies are decision rules that assign treatments based on whether an observable characteristic exceeds a certain threshold. They are widespread across multiple domains, including welfare programs, taxation, and clinical medicine. This paper examines the problem of designing threshold policies using experimental data, when the goal is to maximize the population welfare. First, I characterize the regret - a measure of policy optimality - of the Empirical Welfare Maximizer (EWM) policy, popular in the literature. Next, I introduce the Smoothed Welfare Maximizer (SWM) policy, which improves the EWM's regret convergence rate under an additional smoothness condition. The two policies are compared by studying how differently their regrets depend on the population distribution, and investigating their finite sample performances through Monte Carlo simulations. In many contexts, the SWM policy guarantees larger welfare than the EWM. An empirical illustration demonstrates how the treatment recommendations of the two policies may differ in practice.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (36)
  1. On the bootstrap of the maximum score estimator. Econometrica 73(4), 1175–1204.
  2. Machine learning and phone data can improve targeting of humanitarian aid. Nature 603(7903), 864–870.
  3. Amemiya, T. (1985). Advanced econometrics. Harvard university press.
  4. Policy learning with observational data. Econometrica 89(1), 133–161.
  5. Banerjee, M. and I. W. McKeague (2007). Confidence sets for split points in decision trees. The Annals of Statistics 35(2), 543–574.
  6. Banerjee, M. and J. A. Wellner (2001). Likelihood ratio tests for monotone functions. Annals of Statistics, 1699–1731.
  7. The benefits and costs of jtpa title ii-a programs: Key findings from the national job training partnership act study. Journal of human resources, 549–576.
  8. The impact of nearly universal insurance coverage on health care utilization: evidence from medicare. American Economic Review 98(5), 2242–2258.
  9. Bootstrap-based inference for cube root asymptotics. Econometrica 88(5), 2203–2219.
  10. Chernoff, H. (1964). Estimation of the mode. Annals of the Institute of Statistical Mathematics 16(1), 31–41.
  11. Aid under fire: Development projects and civil conflict. American Economic Review 104(6), 1833–1856.
  12. Groeneboom, P. and J. A. Wellner (2001). Computing chernoff’s distribution. Journal of Computational and Graphical Statistics 10(2), 388–400.
  13. Targeting impact versus deprivation. Technical report, National Bureau of Economic Research.
  14. Hirano, K. and J. R. Porter (2009). Asymptotics for statistical treatment rules. Econometrica 77(5), 1683–1701.
  15. Horowitz, J. L. (1992). A smoothed maximum score estimator for the binary response model. Econometrica: journal of the Econometric Society, 505–531.
  16. Targeting high ability entrepreneurs using community information: Mechanism design in the field. American Economic Review 112(3), 861–98.
  17. Kamath, P. S. and W. R. Kim (2007). The model for end-stage liver disease (meld). Hepatology 45(3), 797–805.
  18. Cube root asymptotics. The Annals of Statistics, 191–219.
  19. Treatment choice with nonlinear regret. arXiv preprint arXiv:2205.08586.
  20. Who should be treated? empirical welfare maximization methods for treatment choice. Econometrica 86(2), 591–616.
  21. Decision trees as partitioning machines to characterize their generalization properties. Advances in Neural Information Processing Systems 33, 18135–18145.
  22. On the bootstrap in cube root asymptotics. Canadian Journal of Statistics 34(1), 29–44.
  23. Manski, C. F. (1975). Maximum score estimation of the stochastic utility model of choice. Journal of econometrics 3(3), 205–228.
  24. Manski, C. F. (2004). Statistical treatment rules for heterogeneous populations. Econometrica 72(4), 1221–1246.
  25. Manski, C. F. (2021). Econometrics for decision making: Building foundations sketched by haavelmo and wald. Econometrica 89(6), 2827–2853.
  26. Manski, C. F. (2023). Probabilistic prediction for binary treatment choice: with focus on personalized medicine. Journal of Econometrics 234(2), 647–663.
  27. Statistical decision theory respecting stochastic dominance. The Japanese Economic Review 74(4), 447–469.
  28. Model selection for treatment choice: Penalized welfare maximization. Econometrica 89(2), 825–848.
  29. Asymptotics in empirical risk minimization. Journal of Machine Learning Research 6(12).
  30. Rai, Y. (2018). Statistical inference for treatment assignment policies. Unpublished Manuscript.
  31. Shigeoka, H. (2014). The effect of patient cost sharing on utilization, health, and risk protection. American Economic Review 104(7), 2152–2184.
  32. Stoye, J. (2012). Minimax regret treatment choice with covariates or with limited validity of experiments. Journal of Econometrics 166(1), 138–156.
  33. Treatment allocation under uncertain costs. arXiv preprint arXiv:2103.11066.
  34. Sun, L. (2021). Empirical welfare maximization with constraints. arXiv preprint arXiv:2103.15298.
  35. Taylor, J. (2003). Corporation income tax brackets and rates, 1909-2002. Statistics of Income. SOI Bulletin 23(2), 284–291.
  36. Fair policy targeting. Journal of the American Statistical Association, 1–14.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now