A Behavioural Research in Finance Seminar
Date/Time: Tuesday 20 June, 14.30-16.00
Venue: Room 4.45, Newcastle University Business School
We welcome Konstantinos Georgaalos, of Lancaster University Management School to present his research: An Experimental Test of the Descriptive and Predictive Power of Dynamic Ambiguity Models
In this paper we report results from an economic experiment where we compare the descriptive and predictive validity 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 in-sample and out-of-sample fit.
Our data show that a dynamic specification of Prospect Theory has the highest expected probability to be the best model for any random subject. Choquet Expected Utility performs best in predicting behaviour, closely followed by Prospect Theory, suggesting that the cost of assuming the wrong specification to predict behaviour is very low compared to the gains from the modelling parsimony that Prospect Theory is offering.