2021
DOI: 10.48550/arxiv.2108.06473
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Evidence Aggregation for Treatment Choice

Takuya Ishihara,
Toru Kitagawa

Abstract: Consider a planner who has to decide whether or not to introduce a new policy to a certain local population. The planner has only limited knowledge of the policy's causal impact on this population due to a lack of data but does have access to the publicized results of intervention studies performed for similar policies on different populations. How should the planner make use of and aggregate this existing evidence to make her policy decision? Building upon the paradigm of 'patient-centered meta-analysis' prop… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
18
0

Year Published

2021
2021
2023
2023

Publication Types

Select...
4
1

Relationship

1
4

Authors

Journals

citations
Cited by 5 publications
(18 citation statements)
references
References 32 publications
0
18
0
Order By: Relevance
“…This framework nests some existing setups of treatment choice, such as limit experiments under parametric models by Hirano and Porter (2009), Gaussian experiments with limited validity by Stoye (2012), and a setup of policy choice based on multiple studies by Ishihara and Kitagawa (2021). 7 One of the essential departures from these setups is that the parameter θ can be infinite dimensional, accommodating nonparametric regression models, for example.…”
Section: Setupmentioning
confidence: 84%
See 4 more Smart Citations
“…This framework nests some existing setups of treatment choice, such as limit experiments under parametric models by Hirano and Porter (2009), Gaussian experiments with limited validity by Stoye (2012), and a setup of policy choice based on multiple studies by Ishihara and Kitagawa (2021). 7 One of the essential departures from these setups is that the parameter θ can be infinite dimensional, accommodating nonparametric regression models, for example.…”
Section: Setupmentioning
confidence: 84%
“…Therefore, the robustness of the nonrandomized minimax regret rule to the error variance and to the parameter space is not a general property. 23 In a special case of this paper's setup, Ishihara and Kitagawa (2021) characterize the decision rule that minimizes the maximum regret within the class of decision rules of the form δ(Y ) = 1{w Y ≥ 0}, where w ∈ R n . Theorem 1 shows that this restricted class contains the minimax regret rule when σ ≥ 2φ(0) ω( 0) ω (0) and does not when σ < 2φ(0) ω( 0) ω (0) .…”
Section: Relation To Existing Resultsmentioning
confidence: 99%
See 3 more Smart Citations