Abstract:
This paper develops a utility model for evaluating lotteries. In estimating utility, risk averse people use an asymmetric loss function. Expected utility is seen as a special case that is a good approximation of the general case in some cases. The model resolves several paradoxes and makes easily falsifiable predictions. When used in hypothesis testing, the model allows researchers to directly specify their attitudes toward risk. The model is advantageous for two reasons. First, it is based on established principles of probability; second, it resolves several well- known paradoxes.