Baillon, Aurélien (2007) Disentangling beliefs and attitudes towards uncertainty in individual decision-making. PhD thesis Sciences économiques, GRID, ENSAM 2007ENAM0026 p.281.

Full text available as: - Thèse_AB.pdf ( 2439 Kb )

Licence: Copyright

Abstract

This dissertation deals with the theory of decision making under uncertainty. It aims at describing and modeling decisions in order to disentangle beliefs and attitudes towards uncertainty. It is made of four main contributions. The first one is theoretical and characterize risk and ambiguity attitude through tradeoff relations. Indeed, expected utility and some of its generalization represents risk and ambiguity aversion through decreasing marginal utility. The tradeoff relation is defined on consequences and allows us to compare concavity of different utility functions under expected utility when probabilities need not be known. The other parts of the dissertation are based on nonexpected utility models including probability weighting functions. The second contribution is an experimental study of how decision makers combine experts’ probability judgments. A new choice-based method is proposed and applied to the comparison of two typical situations. In the first one, the experts give an imprecise evaluation of the risk and in the second one, they disagree and each of them gives his/her own evaluation. The third work is based on uniform sources of uncertainty, i.e. set of events that pertain to a similar mechanism of uncertainty and on which a probability measure exists. In an experiment, such probabilities are elicited. Then willingness-to-bet on events having the same subjective probabilities but from different sources are obtained. The last work is specifically dedicated to the technique for eliciting subjective probability. It observes that this technique is more robust than several well-known techniques and provides new evidence regarding its feasibility.

Statistiques de consultation

Repository Staff Only: edit this item