This paper examines estimation issues associated with multivariate tests of asset pricing. Two issues are considered: (1) the constraint that the sample size (<italic>N</italic>) must be less than the time series (<italic>T</italic>), and (2) the relative effect on power of using the multivariate statistic versus a univariate...

The robustness of the multivariate tests of Michael R. Gibbons, Stephen A. Ross, and Jay Shanken (1986) to nonnormalities in the residual covariance matrix is examined. After considering the relative performance of various tests of normality, simulation techniques are used to determine the...

An intertemporal general equilibrium model relates financial asset returns to movements in aggregate output. The model is a standard neoclassical growth model with serial correlation in aggregate output. Changes in aggregate output lead to attempts by agents to smooth consumption, which affects...

When estimating market model betas using random coefficient methods, the rather fine distinction between significance or insignificance, as argued in recent studies, overlooks two important factors. First, the maximum likelihood method has not been tested in comparison to the generalized least...