摘要：The exponential utility function for money has long attracted attention from theorists because it exhibits nonincreasing absolute risk aversion. Also, under certain conditions, it generates an expected utility function that is maximizable in a quadratic program. However, this functional form presents estimation problems. Logarithmic transformation of an exponential utility function does not conform to the Von Neumann-Morgenstern axioms. Hence, it cannot be used as a basis for best fit in statistical analysis. A criterion is described that can be used to select a best-fit exponential utility function, and its application in grower utility analysis is demonstrated.
