We propose a simple model of equilibrium asset pricing in which there are differences in the amounts of information available for developing inferences about the returns parameters of alternative securities. In contrast with earlier work, we show that parameter uncertainty, or estimation risk, can have an effect upon market equilibrium. Under reasonable conditions, securities for which there is relatively little information are shown to have relatively higher systematic risk when that risk is properly measured, ceteris paribus. The initially very limited model is shown to be robust with respect to relaxation of a number of its principal assumptions. We provide theoretical support for the empirical examination of at least three proxies for relative information: period of listing, number of security returns observations available, and divergence of analyst opinion.

This paper examines properties of daily stock returns and how the particular characteristics of these data affect event study methodologies. Daily data generally present few difficulties for event studies. Standard procedures are typically well-specified even when special daily data characteristics are ignored. However, recognition of autocorrelation in daily excess returns and changes in their variance conditional on an event can sometimes be advantageous. In addition, tests ignoring cross-sectional dependence can be well-specified and have higher power than tests which account for potential dependence.