Mutual Fund Survivorship

摘要: This paper offers a comprehensive study of survivorship issues, in the context mutual fund research, using data set Carhart (1997). We find that funds our sample disappear primarily because multi-year poor performance. Then we demonstrate analytically this survival rule typically causes survivor bias average performance to increase length period, though it is possible construct counterexamples. In data, strong positive relation between and period length. The economically small at 17 basis points per annum for one-year samples, but significantly larger one percent samples longer than fifteen years. also evidence persistence and, consistent with presence multi-period rule, weakened by bias. Finally, explain how characteristics can be affected use survivor-only show magnitudes biases slope coefficients are large size, expenses, turnover load fees sample. Because issues relevant many sets used finance, analysis has potential applications areas financial economics beyond just research.