The possibility of Bayesian learning in repeated games
In infinitely repeated games, Nachbar (1997, 2005) has shown that Bayesian learning of a restricted strategy set is inconsistent; the beliefs required to learn any element of such a set will lead best responses to lie outside of it in most games. But I establish here that Nash convergence of Bayesia...
| Главный автор: | |
|---|---|
| Формат: | Journal article |
| Язык: | English |
| Опубликовано: |
Elsevier BV
2022
|
| _version_ | 1826308960144064512 |
|---|---|
| author | Norman, TWL |
| author_facet | Norman, TWL |
| author_sort | Norman, TWL |
| collection | OXFORD |
| description | In infinitely repeated games, Nachbar (1997, 2005) has shown that Bayesian learning of a restricted strategy set is inconsistent; the beliefs required to learn any element of such a set will lead best responses to lie outside of it in most games. But I establish here that Nash convergence of Bayesian learning requires only that optimal play (rather than any possible play) is learnable, and an appropriately modified notion of learnability is consistent in many of the games to which Nachbar's result applies. This means that rational learning of equilibrium is possible in an important class including coordination games, which I illustrate with two examples of positive learning results. |
| first_indexed | 2024-03-07T07:27:08Z |
| format | Journal article |
| id | oxford-uuid:46a90a8b-d442-4a2f-a98e-bc5b40924e17 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T07:27:08Z |
| publishDate | 2022 |
| publisher | Elsevier BV |
| record_format | dspace |
| spelling | oxford-uuid:46a90a8b-d442-4a2f-a98e-bc5b40924e172022-11-30T16:18:17ZThe possibility of Bayesian learning in repeated gamesJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:46a90a8b-d442-4a2f-a98e-bc5b40924e17EnglishSymplectic ElementsElsevier BV2022Norman, TWLIn infinitely repeated games, Nachbar (1997, 2005) has shown that Bayesian learning of a restricted strategy set is inconsistent; the beliefs required to learn any element of such a set will lead best responses to lie outside of it in most games. But I establish here that Nash convergence of Bayesian learning requires only that optimal play (rather than any possible play) is learnable, and an appropriately modified notion of learnability is consistent in many of the games to which Nachbar's result applies. This means that rational learning of equilibrium is possible in an important class including coordination games, which I illustrate with two examples of positive learning results. |
| spellingShingle | Norman, TWL The possibility of Bayesian learning in repeated games |
| title | The possibility of Bayesian learning in repeated games |
| title_full | The possibility of Bayesian learning in repeated games |
| title_fullStr | The possibility of Bayesian learning in repeated games |
| title_full_unstemmed | The possibility of Bayesian learning in repeated games |
| title_short | The possibility of Bayesian learning in repeated games |
| title_sort | possibility of bayesian learning in repeated games |
| work_keys_str_mv | AT normantwl thepossibilityofbayesianlearninginrepeatedgames AT normantwl possibilityofbayesianlearninginrepeatedgames |