• The two major conclusions of the Sharpe-Lintner CAPM are that (1) the market portfolio is a mean-variance efficient portfolio; and (2) the excess return of each security is proportional to its beta.
• The “market portfolio” includes all securities in the market.
• The beta (β) in CAPM is estimated using regression analysis using historical data on observed returns for a security (response variable) and observed returns for the market (explanatory variable).
• The Roy CAPM differs from the Sharpe-Lintner CAPM only in its assumption concerning the investment constraint imposed by investors. More specifically, it assumes that each investor can short securities.
• Confusion regarding the CAPM involves (1) the failure to distinguish between the following two statements: the market is efficient in that each participant has correct beliefs and uses them to their advantage on the one hand, and the market portfolio is a mean-variance efficient portfolio on the other hand; (2) belief that CAPM investors get paid for bearing nondiversifiable risk; and (3) failure to distinguish between the beta in Sharpe’s one-factor model of covariance (1963 beta) and that in Sharpe’s CAPM (1964 beta).