Yes, CAPM is dead

This paper proves CAPM, as derived by Sharpe, Lintner, and Mossin, faces either a serious problem of endogeneity or of circularity, or both. CAPM thus cannot serve the purposes originally intended for it. The expectation of an unconditional random variable is a constant parameter as set by the density function and should not be affected by other factors. However, CAPM asserts the expected excess rate of return on an asset depends on its beta, which in turn depends on the covariance between the asset's return and the market portfolio's rate of return. In effect, the expected rate of return for an asset must be assumed or known to form the covariance matrix necessary for CAPM. It follows that beta depends on the expected rate of return for an asset, not vice versa, exactly opposite of the presumed relationship. This paper also addresses the validity of the market index model used in empirical studies.