Comparing Consumption-Based Asset-Pricing models

We make use of a recently developed method to estimate the intertemporal marginal rate of substitution consistent with the fluctuations of asset return data from the Toronto Stock Exchange. These estimates are then used to evaluate various parametric specifications for preferences often used in empirical studies of consumption and asset returns. In contrast to existing studies, we are able to perform a formal statistical comparison of these models. We consider six extensions of the usual power utility model, and we find that none can be said to be a demonstrable improvement on the standard model.