Titolo
Learning to drink beer by mistakePeriodico
European University Institute of Badia Fiesolana (Fi). Department of Economics - Working papersAnno:
1994 - Fascicolo:
30 - Pagina iniziale:
1 - Pagina finale:
15This paper considers the possibility of using adaptive learning and mistakes to select a unique equilibrium in signaling games. It is assumed that players are boundedly rational in that they use an adaptive rule to update their beliefs. Moreover, they sometimes make mistakes and choose an action at random. By computer simulation it is shown that, when players do not make mistakes, the equilibrium selected depends on the initial distribution of beliefs. When the probability of mistakes is positive the learning dynamics selects for
Kohlberg-Mertens’ stability concept.
Testo completo:
http://hdl.handle.net/1814/509Esportazione dati in Refworks (solo per utenti abilitati)
Record salvabile in Zotero