This paper proposes a bayesian methodology to treat the who’s who problem arising in individual level data sets such as patent data. We assess the usefullness of this methodology on the set of all French inventors appearing on EPO applications from 1978 to 2003.

Often, perfect bayesian equilibrium is loosely defined by stating that players should be sequentially rational given some beliefs in which Bayes rule is applied whenever possible". We show that there are games in which it is not clear what whenever possible" means. Then, we provide a simple...

Often, perfect bayesian equilibrium is loosely defined by stating that players should be sequentially rational given some beliefs in which Bayes rule is applied “whenever possible”. We show that there are games in which it is not clear what “whenever possible” means. Then, we provide a...