Date Added: Jul 2011
The authors propose two metrics for asset pricing models and apply them to representative agent models with recursive preferences, habits, and jumps. The metrics describe the pricing kernel's dispersion (the entropy of the title) and dynamics (time dependence, a measure of how entropy varies over different time horizons). They show how each model generates entropy and time dependence and compare their magnitudes to estimates derived from asset returns. This exercise - and transparent loglinear approximations - clarifies the mechanisms underlying these models.