Strategic Experimentation with Exponential Bandits.
This paper studies a game of strategic experimentation with two-armed bandits whose risky arm might yield a payoff only after some exponentially distributed random time. Because of free-riding, there is an inefficiently low level of experimentation in any equilibrium where the players use stationary...
| Main Authors: | , , |
|---|---|
| Format: | Working paper |
| Language: | English |
| Published: |
Department of Economics (University of Oxford)
2003
|
| _version_ | 1826276760885395456 |
|---|---|
| author | Cripps, M Keller, G Rady, S |
| author_facet | Cripps, M Keller, G Rady, S |
| author_sort | Cripps, M |
| collection | OXFORD |
| description | This paper studies a game of strategic experimentation with two-armed bandits whose risky arm might yield a payoff only after some exponentially distributed random time. Because of free-riding, there is an inefficiently low level of experimentation in any equilibrium where the players use stationary Markovian strategies with posterior beliefs as the state variable. After characterizing the unique symmetric Markovian equilibrium of the game, which is in mixed strategies, we construct a variety of pure-strategy equilibria. There is no equilibrium where all players use simple cut-off strategies. Equilibria where players switch finitely often between the roles of experimenter and free-rider all lead to the same pattern of information acquisition; the efficiency of these equilibria depends on the way players share the burden of experimentation among them. In equilibria where players switch roles infinitely often, they can acquire an approximately efficient amount of information, but the rate at which it is acquired still remains inefficient; moreover, the expected payoff of an experimenter exhibits the novel feature that it rises as players become more pessimistic. Finally, over the range of beliefs where players use both arms a positive fraction of the time, the symmetric equilibrium is dominated by any asymmetric one in terms of aggregate payoffs. |
| first_indexed | 2024-03-06T23:18:42Z |
| format | Working paper |
| id | oxford-uuid:6802ab02-0d92-4f11-8cf5-8699e25ad6a6 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-06T23:18:42Z |
| publishDate | 2003 |
| publisher | Department of Economics (University of Oxford) |
| record_format | dspace |
| spelling | oxford-uuid:6802ab02-0d92-4f11-8cf5-8699e25ad6a62022-03-26T18:42:05ZStrategic Experimentation with Exponential Bandits.Working paperhttp://purl.org/coar/resource_type/c_8042uuid:6802ab02-0d92-4f11-8cf5-8699e25ad6a6EnglishDepartment of Economics - ePrintsDepartment of Economics (University of Oxford)2003Cripps, MKeller, GRady, SThis paper studies a game of strategic experimentation with two-armed bandits whose risky arm might yield a payoff only after some exponentially distributed random time. Because of free-riding, there is an inefficiently low level of experimentation in any equilibrium where the players use stationary Markovian strategies with posterior beliefs as the state variable. After characterizing the unique symmetric Markovian equilibrium of the game, which is in mixed strategies, we construct a variety of pure-strategy equilibria. There is no equilibrium where all players use simple cut-off strategies. Equilibria where players switch finitely often between the roles of experimenter and free-rider all lead to the same pattern of information acquisition; the efficiency of these equilibria depends on the way players share the burden of experimentation among them. In equilibria where players switch roles infinitely often, they can acquire an approximately efficient amount of information, but the rate at which it is acquired still remains inefficient; moreover, the expected payoff of an experimenter exhibits the novel feature that it rises as players become more pessimistic. Finally, over the range of beliefs where players use both arms a positive fraction of the time, the symmetric equilibrium is dominated by any asymmetric one in terms of aggregate payoffs. |
| spellingShingle | Cripps, M Keller, G Rady, S Strategic Experimentation with Exponential Bandits. |
| title | Strategic Experimentation with Exponential Bandits. |
| title_full | Strategic Experimentation with Exponential Bandits. |
| title_fullStr | Strategic Experimentation with Exponential Bandits. |
| title_full_unstemmed | Strategic Experimentation with Exponential Bandits. |
| title_short | Strategic Experimentation with Exponential Bandits. |
| title_sort | strategic experimentation with exponential bandits |
| work_keys_str_mv | AT crippsm strategicexperimentationwithexponentialbandits AT kellerg strategicexperimentationwithexponentialbandits AT radys strategicexperimentationwithexponentialbandits |