6I8 JOURNAL OF POLITICAL ECONOMY TABLE 1 PORTFOLIO FORMATION,ESTIMATION,AND TESTING PERIODS PERIODS 2 Portfolio formation period... 1926-29 1927-33 1931-37 1935-41 1939-45 Initial estimation period ..... 1930-34 1934-38 1938-42 1942-46 1946-50 Testing period 1935-38 1939-42 1943-46 1947-50 1951-54 No.of securities available ... 710 779 804 908 1,011 No.of securities meeting data requirement.......... 435 576 607 704 751 s(B).Estimates of B for portfolios are indeed more precise than those for individual securities. Nevertheless,it is interesting to note that if the disturbances e in (8) were independent from security to security,the relative increase in the precision of the B obtained by using portfolios rather than individual securities would be about the same for all portfolios.We argue in the Appendix,however,that the results from (10)imply that the t in (8)are interdependent,and the interdependence is strongest among high-B secu- rities and among low-B securities.This is evident in table 2:The ratios s(p)/5.1)are always highest at the extremes of the Bp.range and lowest for B.1close to 1.0.But it is important to emphasize that since these ratios are generally less than .33,interdependence among the t of different securities does not destroy the value of using portfolios to reduce the dispersion of the errors in estimated B's. Finally,all the tests of the two-parameter model are predictive in the sense that the explanatory variables Br.t-1and 3.1()in (10)are com- puted from data for a period prior to the month of the returns,the Rot,on which the regression is run.Although we are interested in testing the two- parameter model as a positive theory-that is,examining the extent to which it is helpful in describing actual return data-the model was initially developed by Markowitz (1959)as a normative theory-that is,as a model to help people make better decisions.As a normative theory the model only has content if there is some relationship between future returns and esti- mates of risk that can be made on the basis of current information. Now that the predictive nature of the tests has been emphasized,to simplify the notation,the explanatory variables in (10)are henceforth referred to as B,B",and p(). V.Results The major tests of the implications of the two-parameter model are in table 3.Results are presented for 10 periods:the overall period 1935-