A Microeconometric Test of Alternative Stochastic Theories of Risky Choice.

We compare the random preference, Fechner, and constant error (or "tremble") approaches to the stochastic modelling of choice under risk. Various combinations of these approaches are used with expected utility and rank-dependent theory to generate a set of econometric models. These are estimated in a random effects framework using experimental data from two samples of 46 subjects who each faced 90 pairwise choice problems. The best fitting model uses the random preferences approach with an added tremble mechanism, in conjunction with rank-dependent theory. As subject gain experience, trembles become less frequent and there is less deviation from expected utility theory.