On the diversification, observability, and measurement of estimation risk

This paper reexamines how risk return relationships are affected by investor uncertainty about the exact parameters of the joint rate of return distribution. We attempt to clarify results relating to three central issues. First, we address the issue of diversification, focusing on an APT, factor model framework. Second, we discuss the observability of estimation risk and describe research experimental designs that should encompass the existence of estimation risk and reveal it in the data. Finally, we suggest exploiting contemporaneous return observations on high and low information securities to aid in the measurement of return parameters for low information securities.