Reviewing the basics of mean–variance portfolio optimization and the capital asset pricing model, this paper discusses the plausibility of some of the underlying assumptions. It is pointed out that a positive in-sample relationship between the expected return of an asset and its covariance with the market portfolio can be a statistical artifact because it can be explained without using any economic arguments. In an empirical analysis of two sets of assets consisting of individual stocks and indices, respectively, no indication of any out-of-sample relationship is found.
