To detect funds with positive or negative performance, the standard approach in the academic literature (., Jensen (1968), Ferson and Schadt (1996), Ferson and Qian (2004)) can be described as follows. The presence of differential performance (positive or negative alphas) is tested for each of the M funds in the population. Then, a conventional significance level γ is set (., 5 percent) and all funds with p-values smaller than γ are said to have significant estimated alphas. Finally, these significant alphas are counted in order to provide an estimate of the number of funds with differential performance
