Solving an empirical puzzle in the capital asset pricing model

A long standing puzzle in the Capital Asset Pricing Model (CAPM) has been the inability of empirical work to validate it. This paper presents a new approach to estimating the CAPM, taking into account the differences between observable and expected returns for risky assets and for the market portfolio of all traded assets, as well as inherent nonlinearities and the effects of excluded variables. Using this approach, we provide evidence that the relation between the observable returns on stock and market portfolios is nonlinear.