An experimental test of the predictive power of dynamic ambiguity models

Abstract

In this paper we report results from an economic experiment where we investigate the predictive performance of dynamic ambiguity models in the gains domain. Representing ambiguity with the aid of a transparent and non-manipulable device (a Bingo Blower) and using two-stage allocation questions, we gather data that allow us to estimate particular parametric forms of the various functionals and compare their relative performance in terms of out-of-sample fit. Our data show that a dynamic specification of Prospect Theory has the best predictive capacity, closely followed by Choquet Expected Utility, while multiple priors theories can predict choice only for a very restricted subset of our subjects.

Iterated expectations under rank‐dependent expected utility and implications for common valuation methods

“…This specification is attractive from a theoretical point of view (model consistency), and it recently also received some empirical support. In an experimental test, Georgalos (2019) runs a horse race between a number of dynamic models of choice and updating. He finds that prospect theory with updating of the probability weighting function is the winner in predicting out of sample behaviour after partial resolution of uncertainty.…”

Dynamic decision making under ambiguity: An experimental investigation

“…Kahneman (1992) Prospect Theory specification. Georgalos (2019) assumes consequentialism and estimates various models and updating rules. Although the present study will use the same data set as in Georgalos (2019), we extend their analysis in two ways: first, we test which of the two axioms is violated by ambiguity non-neutral agents.…”