Disentangling beliefs and attitudes towards uncertainty in individual decision making.

Authors
Publication date
2007
Publication type
Thesis
Summary The thesis deals with the theory of individual decision under uncertainty. It aims at understanding, describing and representing decisions, by differentiating what comes from the decision maker's beliefs from what is related to his attitude towards uncertainty. It is mainly composed of four complementary contributions. The first one is theoretical and characterizes the attitude towards risk and ambiguity through the use of trade-off relations concerning the consequences, in the line of expected utility models, representing the aversion to risk and/or ambiguity through the decrease of the marginal utility. The rest of the thesis is based on models generalizing expected utility where probabilities are transformed by the decision maker. The second contribution is then experimental and focuses on the decision-maker who has expert opinions indicating the risk incurred. A method is proposed, based on the observation of choices, to study how the decision-maker combines the opinions at his disposal. This method is applied to compare situations where experts give an imprecise assessment of the risk with situations where their assessments of the risk incurred are conflicting. The third work introduces the concept of a uniform source of uncertainty, i.e. a set of events generated by the same uncertainty mechanism and for which there is a subjective probability measure. An experiment is conducted in which such subjective probabilities are obtained. The willingness of individuals to bet on events of equivalent (subjective) probability but coming from different sources is then studied. The last contribution goes back to the method of obtaining subjective probabilities and compares it (theoretically) to other methods. Its feasibility and limitations are then studied in a new experiment.
Topics of the publication
  • ...
  • No themes identified
Themes detected by scanR from retrieved publications. For more information, see https://scanr.enseignementsup-recherche.gouv.fr