PREFERENCE FOR GRADUAL RESOLUTION OF UNCERTAINTY

Analyses of preference for the timing of uncertainty resolution usuall y assumes all uncertainty to resolve in one point in time. More realis tically, uncertainty should be modelled to resolve gradually over time . Kreps and Porteus (1978) have introduced an axiomatically based mode l of time preference which can explain preferences for gradual uncerta inty resolution. This paper presents an experimental test of the Kreps -Porteus model. We derive implications of the model relating preferenc es for gradual and one-time resolving lotteries. Our data do not suppo rt the Kreps-Porteus model but show that some of the behaviour observe d may be explained by similarity heuristics.