Footnote:
Nach Informationen von SSRN wurde die ursprüngliche Fassung des Dokuments August 3, 2022 erstellt
Description:
We experimentally study agents’ preferences for ambiguity resolution in dynamic environments. We theoretically demonstrate that the three most popular recursive models of ambiguity aversion make different predictions regarding agents’ preferences for the timing and graduality of ambiguity resolution. We test these models in an extension of the original Ellsberg experiment. Our results show the existence of preference for late and gradual resolution of ambiguity. In addition, there is a strong interdependence between the ambiguity attitude and the preference for the timing and graduality of the resolution of ambiguity. The smooth model of ambiguity (Klibanoff, Marinacci, and Mukerji, 2009) is the only model that is flexible enough to accommodate the observed preferences. In contrast, the $\alpha-$maxmin model of Ghirardato, Maccheroni, and Marinacci (2004) predicts indifference to the timing and graduality of ambiguity resolution, while the multiplier preferences of Hansen and Sargent (2001) cannot sufficiently separate ambiguity aversion from temporal attitudes. This interdependence of ambiguity and timing is of interest both conceptually and practically—especially for economists using these models in applications