复旦大学投资学课件Section3 Returns equity mutual funds.pptVIP

Returns from investing in equity mutual funds 1971 to 1991 Burton G. Malkiel JOF 1995 Introduction Early 1970s, EMH accepted By the early 1980s, several cracks: returns are not independent and correlated over time. The predictability of returns Most efficiency test are joint tests Eugene Fama:”Sequels are rarely as good as the originals.” Jensen(1968): performance of mutual funds was inferior to randomly selected portfolios with equivalent risk Henriksson(1984): fund managers have enough private information to offset expenses. hot hand phenomenon Section 1: data Section 2: performance of the equity mutual funds Section 3: hot hand phenomenon Section 4: simulate investment strategies Section 5: relation between returns and expenses conclude 1 Survivorship bias and the data set employed Today’s investors are not interested in the funds that no longer exist, which creates the biases in the figures. High returns: the funds failed have been dropped out of the sample Tendency of more successful funds to survive, overstated. “incubator” funds Funds data 1971-1991: all general equity funds sold to the public. Quarterly total returns. Some survivorship in table 1 The differences is big 2. A closer look at performance The t is only -0.21, so it is indistinguishable from 0. even surviving funds do not produce excess returns for investors after expenses. The numbers of positive alpha and negative alpha are approximately equal. Table 3 Grossman and Stiglitz(1980): positive alphas for pre-expense returns confirm that mutual funds do earn gross returns to cover expenses But positive alphas are small and insignificant. EGDH(1993): inappropriate benchmarks, if corrects for the non-SP stocks, the positive alphas disappear. The 1980s: small stocks tended to underperform the SP stock index. 10-year sample limited There is relationship between returns and betas in mutual funds. The betas are stable Over 20 year period 1971 to 1991, the relationships between betas and total ret