Bray Theory Workshop

This paper builds a procedure to partially identify and estimate the share of households that conform to various models of decision making under uncertainty – such as probability weighting, reference dependent loss aversion, and standard expected utility – and uses the estimated decision types and preferences to measure welfare implications of salient interventions in insurance markets.  The procedure leverages advances in random set theory to overcome typical identification concerns that arise when households' decisions are consistent with multiple types of rational behavior.  It is computationally tractable and informative about both the underlying preference structure and the shares of various decision types.