Reinforcing RCTs with Multiple Priors while Learning about External Validity
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| 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 |