SMU SOE Seminar Series (April 3, 2025): Decision Making Under Multidimensional Risk

TOPIC:

ABSTRACT

Choice alternatives are often multidimensional and risky, but how to model a decision maker's evaluation of them is unclear. Three popular approaches are in sharp contrast: One aggregates all dimensions before evaluating risk, one does it reversely, and one evaluates each dimension's risk recursively. We characterize a hierarchical expected utility model that generalizes these approaches. The decision maker's preference reveals how she brackets and orders the dimensions, based on which she evaluates risk recursively. We analyze the model's uniqueness properties and characterize several special cases. We study the model's implications in contexts of group inequality, multisource income, and time lotteries.

Click here to view the CV.
Click here to view the paper.