Giusti, Giuliano ; Zagonari, Fabio
(1997)
From a Regret to an Expected Utility Model: a learning process.
Bologna:
Dipartimento di Scienze economiche DSE,
p. 11.
DOI 10.6092/unibo/amsacta/5026.
In: Quaderni - Working Paper DSE
(284).
ISSN 2282-6483.
Full text disponibile come:
Anteprima |
Documento di testo(pdf)
Licenza: Creative Commons Attribution Non-commercial 3.0 (CC BY-NC 3.0) Download (152kB) | Anteprima |
Abstract
This paper identifies in a feed-forward neural network the mathematical algorithm which can catch the learning process highlighted by econometric works that makes people assess the satisfaction arising in each single contingency so that they are better depicted in their decision making by an Expected Utility rather than by a Regret Model. Evidence from experimental economics are also accounted for, since the network does not manage to extrapolate the former from the latter model when probabilities are extreme.
Abstract