%0 Journal Article
%T Hierarchical Bayesian Analysis of Biased Beliefs and Distributional Other-Regarding Preferences
%A Ozan Aksoy
%A Jeroen Weesie
%J Games
%D 2013
%I MDPI AG
%R 10.3390/g4010066
%X This study investigates the relationship between an actor¡¯s beliefs about others¡¯ other-regarding (social) preferences and her own other-regarding preferences, using an ¡°avant-garde¡± hierarchical Bayesian method. We estimate two distributional other-regarding preference parameters, ¦Á and ¦Â, of actors using incentivized choice data in binary Dictator Games. Simultaneously, we estimate the distribution of actors¡¯ beliefs about others ¦Á and ¦Â, conditional on actors¡¯£¿own ¦Á and ¦Â, with incentivized belief elicitation. We demonstrate the benefits of the Bayesian method compared to it¡¯s hierarchical frequentist counterparts. Results show a positive association between an actor¡¯s own (¦Á; ¦Â ) and her beliefs about average(¦Á; ¦Â) in the population. The association between own preferences and the variance in beliefs about others¡¯ preferences in the population, however, is curvilinear for ¦Á and insignificant for ¦Â. These results are partially consistent with the cone effect [1,2] which is described in detail below. Because in the Bayesian-Nash equilibrium concept, beliefs and own preferences are assumed to be independent, these results cast doubt on the application of the Bayesian-Nash equilibrium concept to experimental data.
%K Bayesian statistics
%K Dictator Game
%K other-regarding preferences
%K beliefs
%U http://www.mdpi.com/2073-4336/4/1/66