Working Paper: NBER ID: w1643
Authors: Sanford J. Grossman; Angelo Melino; Robert J. Shiller
Abstract: The consumption based asset pricing model predicts that excess yields are determined in a fairly simple way by the market's degree of relative risk aversion and by the pattern of covariances between percapita consumption growth and asset returns. Estimation and testingis complicated by the fact that the model's predictions relate to the instantaneous flow of consumption and point-in-time asset values, but only data on the integral or unit average of the consumption flow is available. In our paper, we show how to estimate the parameters of interest consistently from the available data by maximum likelihood. We estimate the market's degree of relative risk aversion and the instantaneous covariances of asset yields and consumption using six different data sets. We also test the model's overidentifying restrictions.
Keywords: Asset Pricing; Consumption; Risk Aversion
JEL Codes: G12; E21
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| asset covariability (C10) | expected returns (G17) |
| time-averaged data (Y10) | risk aversion parameters (D81) |
| risk aversion (D81) | expected yields (Q47) |
| covariance with consumption changes (D11) | expected excess return (G17) |
| time averaging effects (C22) | biases in estimating risk aversion (D81) |