<> "The repository administrator has not yet configured an RDF license."^^ . <> . . "Decision Making under Ambiguity"^^ . "This thesis deals with decision making under ambiguity. Ambiguity refers to situations in which probabilities for uncertain events are partially known. Ambiguitysensitive behavior, as manifested in Ellsberg-type experiments, is today a widely studied phenomenon, then, first of all, there is an ample empirical evidence confirming ambiguity-sensitive behavior, and second, it cannot be explained by the subjective expected utility theory. The goal of this thesis is to explore further the nature of ambiguity-sensitive behavior in the context of static, dynamic and interpersonal decision problems. For this purpose we apply experimental as well as formal methods. Our first investigation focuses on static decision problems. We test the descriptive validity of the widely accepted methodology used to formalize the notion of different ambiguity attitudes, namely, that ambiguity-averse subjects are randomization-loving, while ambiguity-loving subjects are randomization-averse. Our experimental data do not support this view. Ambiguity-averse subjects are more likely to be randomizationneutral rather than randomization-loving. Furthermore, we also observe a considerable number of ambiguity-averse subjects who exhibit a contempt for randomization. These observations suggest that ambiguity models which do not exogenously assume a specific relationship between ambiguity and randomization attitudes would be better suited to describe real behavior in the presence of ambiguity. Next, we focus on dynamic decision problems. In a dynamic version of the classical 3-color experiment of Ellsberg (1961) we test whether subjects behave consistently with either dynamic consistency or consequentialism. We find that more subjects act in line with consequentialism rather than with dynamic consistency and that this result is even stronger among ambiguity-averse subjects. This evidence can be seen as support for theories of updating ambiguity-sensitive preferences which maintain consequentialism and relax dynamic consistency. Furthermore, we find additional violation of the subjective expected utility theory in the dynamic experiment. Several subjects who are classified as ambiguity-neutral in the static choice situation do not exhibit Bayesian behavior in the dynamic extension. They violate either dynamic consistency or conse quentialism. Therefore, the dynamic version of the 3-color experiment can also be seen as a tool to test Bayesianism and to make the observation from static experiment more robust. In the second part we continue to study dynamic choice problems, but constrain the analysis to the class of Choquet expected utility preferences of Schmeidler (1989). We argue that this class of preferences has a very attractive feature. Namely, it is possible to characterize dynamic properties of Choquet preferences from a static point of view by constraining the analysis to a fixed collection of events. Assuming consequentialism, we show that Choquet expected utility preferences respect dynamic consistence on a fixed collection of events if and only if these events are unambiguous in the sense of Nehring (1999). Accordingly, one can apply the same techniques used in expected utility theory to solve optimization problems, e.g. backward induction. In the last part of this thesis we apply the Choquet expected utility theory to interpersonal decision problems. We show that, unlike in the subjective expected utility theory of Savage, asymmetric information matters in presence of ambiguity and can explain differences in commonly known decisions. Under the assumption of common capacity distribution it is shown that whenever agents’ private information partitions are made up of unambiguous events in the sense of Nehring (1999) then it is impossible that the agents disagree on commonly known decisions, whatever these decisions are, whether conditional beliefs or conditional expectations. Consequently, the possibility of speculative trade is precluded only if private information is made up of unambiguous events in this peculiar sense. Even a small departure from that notion of unambiguous events creates profitable trade opportunities due to differences in agents’ private and ambiguous information. We conclude that the presence of ambiguous private information provides an intuitive explanation for the existence of widely observed speculative activities. Ellsberg, D. (1961): “Risk, Ambiguity, and the Savage Axioms,” Quarterly Journal of Economics, 75, 643–669. Nehring, K. (1999): “Capacities and Probabilistic Beliefs: A Precarious Coexistence,” Mathematical Social Science, 38, 197–213. Schmeidler, D. (1989): “Subjective Probability and Expected Utility without Additivity,” Econometrica, 57, 571–587. "^^ . "2010" . . . . . . . . "Adam"^^ . "Dominiak"^^ . "Adam Dominiak"^^ . . . . . . "Decision Making under Ambiguity (PDF)"^^ . . . "Dissertation_20.pdf"^^ . . . "Decision Making under Ambiguity (Other)"^^ . . . . . . "indexcodes.txt"^^ . . . "Decision Making under Ambiguity (Other)"^^ . . . . . . "lightbox.jpg"^^ . . . "Decision Making under Ambiguity (Other)"^^ . . . . . . "preview.jpg"^^ . . . "Decision Making under Ambiguity (Other)"^^ . . . . . . "medium.jpg"^^ . . . "Decision Making under Ambiguity (Other)"^^ . . . . . . "small.jpg"^^ . . "HTML Summary of #12089 \n\nDecision Making under Ambiguity\n\n" . "text/html" . . . "330 Wirtschaft"@de . "330 Economics"@en . .