Survey_GM_2006

Abstract

Our model has two dates. At date 1, the agent makes a decision under risk. Once the decision has been made, he forms subjective beliefs about the final outcome. These subjective beliefs can diger from the objective probability distribution of the final payog. He extracts pleasure from savoring this prospect. At date 2, the agent observes the payog, which is a function of his date-1 decision and of the realized state of nature. We consider a set of lotteries with fixed support {f1> f2> ===> fV}> where V is the number of states of nature and fv is the real-valued lottery payoff in state s. The structure of our model is such that all subjective beliefs S yielding the same subjective certainty equivalent generate the same intertemporal welfare Y . This implies that the optimal subjective beliefs are undeterminable.