Reinforcing RCTs with Multiple Priors while Learning about External Validity

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Finan, Frederico, Pouzo, Demian
Natura: Preprint
Pubblicazione: 2021
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914959101788160
author Finan, Frederico
Pouzo, Demian
author_facet Finan, Frederico
Pouzo, Demian
contents This paper introduces a framework for incorporating prior information into the design of sequential experiments. These sources may include past experiments, expert opinions, or the experimenter's intuition. We model the problem using a multi-prior Bayesian approach, mapping each source to a Bayesian model and aggregating them based on posterior probabilities. Policies are evaluated on three criteria: learning the parameters of payoff distributions, the probability of choosing the wrong treatment, and average rewards. Our framework demonstrates several desirable properties, including robustness to sources lacking external validity, while maintaining strong finite sample performance.
format Preprint
id arxiv_https___arxiv_org_abs_2112_09170
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle Reinforcing RCTs with Multiple Priors while Learning about External Validity
Finan, Frederico
Pouzo, Demian
Econometrics
Statistics Theory
Methodology
This paper introduces a framework for incorporating prior information into the design of sequential experiments. These sources may include past experiments, expert opinions, or the experimenter's intuition. We model the problem using a multi-prior Bayesian approach, mapping each source to a Bayesian model and aggregating them based on posterior probabilities. Policies are evaluated on three criteria: learning the parameters of payoff distributions, the probability of choosing the wrong treatment, and average rewards. Our framework demonstrates several desirable properties, including robustness to sources lacking external validity, while maintaining strong finite sample performance.
title Reinforcing RCTs with Multiple Priors while Learning about External Validity
topic Econometrics
Statistics Theory
Methodology
url https://arxiv.org/abs/2112.09170