TY - UNPB

T1 - Dynamic decision making under ambiguity

T2 - a portfolio choice experiment

AU - Georgalos, Konstantinos

PY - 2016/4/7

Y1 - 2016/4/7

N2 - Neoclassical economic theory assumes that when agents tackle dynamic decisions under ambiguity, preferences are represented by Expected Utility and prior beliefs are updated according to Bayes rule, upon the arrival of partial information. Nevertheless, when one considers non-neutral ambiguity attitudes, either the axiom of dynamic consistency or of consequentialism should be relaxed. We report the results of a new experiment, designed toinvestigate how people behave in a dynamic choice problem under ambiguity, where decisions are made both before and after the resolution of some uncertainty. We study which of the two rationality axioms people violate, along with the question of whether this violation is part of a conscious planning strategy or not. The combination of the two, allows us to classify subjects to three behavioural types: resolute, naïve and sophisticated. Using data from a portfolio choice experiment where ambiguity is represented in a transparent andnon-manipulable way, we cannot reject the hypothesis of Bayesian updating for half of our experimental population. For ambiguity non-neutral subjects, we find that the majority are sophisticated, a few are naïve and few are resolute.

AB - Neoclassical economic theory assumes that when agents tackle dynamic decisions under ambiguity, preferences are represented by Expected Utility and prior beliefs are updated according to Bayes rule, upon the arrival of partial information. Nevertheless, when one considers non-neutral ambiguity attitudes, either the axiom of dynamic consistency or of consequentialism should be relaxed. We report the results of a new experiment, designed toinvestigate how people behave in a dynamic choice problem under ambiguity, where decisions are made both before and after the resolution of some uncertainty. We study which of the two rationality axioms people violate, along with the question of whether this violation is part of a conscious planning strategy or not. The combination of the two, allows us to classify subjects to three behavioural types: resolute, naïve and sophisticated. Using data from a portfolio choice experiment where ambiguity is represented in a transparent andnon-manipulable way, we cannot reject the hypothesis of Bayesian updating for half of our experimental population. For ambiguity non-neutral subjects, we find that the majority are sophisticated, a few are naïve and few are resolute.

KW - Ambiguity

KW - Subjective Beliefs

KW - Dynamic Consistency

KW - Consequentialism

KW - Portfolio Choice

KW - Experiment

M3 - Working paper

T3 - Economics Working Paper Series

BT - Dynamic decision making under ambiguity

PB - Lancaster University, Department of Economics

CY - Lancaster

ER -