Minimax regret treatment choice with finite samples

By Jörg Stoye. 2009

195 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/279e0f66ee31adaefdf6cc47503e7efa79e89f4e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Statistics and Probability

Citations in the corpus

Estimation of Heterogeneous Treatment Effects from Randomized Experiments, with Application to the Optimal Planning of the Get-Out-the-Vote Campaign (2011, methodology): “In addition, the strategies that meet the minimax regret criterion can include no-data rules that do not depend on the data at all (Stoye 2009).”
Confounding-Robust Policy Improvement (2018, background): “Stoye (2009, 2012) consider minimax regret from a decision-theoretic point of view, where a closed form is available under limiting asymptotic assumptions on an experimental sampling design generating treatment assignments.”
Statistical treatment choice based on asymmetric minimax regret criteria (2009, methodology): “Exact mimimax regret results were obtained for these problems by Stoye (2009) and Schlag (2007).”
Inferring Welfare Maximizing Treatment Assignment Under Budget Constraints (2012, background): “This problem has been analyzed in the literature as a statistical decision problem with finite samples under ambiguity but with no budget constraint (c.f., Dehejia (2001), Manski (2004, 2005), Stoye (2009), Hirano and Porter (2009) and Tetenov (2008)).”
Impossibility Results for Nondifferentiable Functionals (2012, background): “Interest often focuses on estimating the average treatment eect 1 0. Recent work on optimal treatment assignment rules by Manski (2004), Dehejia (2005), Stoye (2006) , Schlag (2006), Tetenov (2007), and Hirano and Porter (2008) has adopted a decision-theoretic approach for choosing the”
BAYESIAN ASPECTS OF TREATMENT CHOICE (2011, background): “Stoye (2009) obtains exact results on minimax regret rules.”
Policy Learning With Observational Data (2017, methodology): “Then, following Manski (2004, 2009), Hirano and Porter (2009), Stoye (2009, 2012) and Kitagawa and Tetenov (2018), we seek guarantees on the regret R (ˆ π ), i.e., the diﬀerence between the expected utility from deploying the learned policy ˆ π over a target population and the best utility that…”
Complementarity and Aggregate Implications of Assortative Matching: A Nonparametric Analysis (2014, unknown): “Our family of reallocations, which we call correlated matching rules, traces a path from the positive to the negative assortative matching allocations.”
Minimax regret treatment choice with covariates or with limited validity of experiments (2012, methodology): “As a practical illustration of these findings, consider an example from Stoye (2009a): There is a binary covariate, ∈ { } say, the decision maker can freely choose the sample design, and both treatments are unknown.”
Efficient Policy Learning (2017, methodology): “In the case where Wi is binary, we follow the existing literature (Hirano and Porter, 2009; Kitagawa and Tetenov, 2018; Manski, 2004; Stoye, 2009), and study interventions that directly specify the treatment level.”
Statistical decisions under ambiguity (2011, background): “…DasGupta and Studden (1991) employ it in regression design, Manski (2004), Hirano and Porter (2005), Schlag (2006), and Stoye (2006) apply it to treatment choice in identified models but with finite samples, and Schlag (2003) brings it to the closely related problem of playing a two-armed bandit.”