Survival, Look-Ahead Bias and the Performance of Hedge Funds

Systemic Risk and Hedge Funds

“…Liang (2000) finds that the annual hedge-fund attrition rate is 8.3% for the 1994-1998 sample period using TASS data, and Baquero, Horst, and Verbeek (2004) find a slightly higher rate of 8.6% for the 1994-2000 sample period. Baquero, Horst, and Verbeek (2004) also find that surviving funds outperform non-surviving funds by approximately 2.1% per year, which is similar to the findings of Hsieh (2000, 2002b) and Liang (2000), and that investment style, size, and past performance are significant factors in explaining survival rates. Many of these patterns are also documented by Liang (2000), Boyson (2002), and Getmansky, Lo, and Mei (2004).…”

“…The authors also find that performance persistence, whenever present, is unrelated to the type of hedge fund strategy. Brown, Goetzmann, Ibbotson, and Ross (1992), Ackermann, McEnally, and Ravenscraft (1999), and Baquero, Horst, and Verbeek (2004) show that survivorship biasthe fact that most hedge-fund databases do not contain funds that were unsuccessful and which went out of business-can affect the first and second moments and cross-moments of returns, and generate spurious persistence in performance when there is dispersion of risk among the population of managers. However, using annual returns of both defunct and currently operating offshore hedge funds between 1989and 1995, Brown, Goetzmann, 13 and Ibbotson (1999 find virtually no evidence of performance persistence in raw returns or risk-adjusted returns, even after breaking funds down according to their returns-based style classifications.…”

“…Hedge-fund survival rates have been studied by Brown, Goetzmann and Ibbotson (1999), Fung and Hsieh (2000), Liang (2000Liang ( , 2001, Bares, Gibson and Gyger (2003), Brown, Goetzmann and Park (2001b), Gregoriou (2002), and Amin and Kat (2003b). Baquero, Horst, and Verbeek (2004) estimate liquidation probabilities of hedge funds and find that they are greatly dependent on past performance. The survival rates of hedge funds have been estimated by Brown, Goetzmann and Ibbotson (1999), Fung and Hsieh (2000), Liang (2000Liang ( , 2001, Brown, Goetzmann and Park (2001a,b), Gregoriou (2002), Amin and Kat (2003b), Bares, Gibson and Gyger (2003), and Getmansky, Lo, and Mei (2004).…”

“…Together, these four categories account for 71.9% of the funds in the Combined database. Figure 1 shows that the relative proportions of the Live and Graveyard 12 For studies attempting to quantify the degree and impact of survivorship bias, see Baquero, Horst, and Verbeek (2004), Brown, Goetzmann, Ibbotson, and Ross (1992), Brown, Goetzmann, and Ibbotson (1999), Brown, Goetzmann, and Park (1997), Carpenter and Lynch (1999), Hsieh (1997b, 2000), Horst, Nijman, and Verbeek (2001), Hendricks, Patel, and Zeckhauser (1997), and Schneeweis and Spurgin (1996).…”

Sifting Through the Wreckage: Lessons From Recent Hedge-Fund Liquidations

“…Brown, Goetzmann, and Park (2001b) show that the probability of liquidation increases with increasing risk, and that funds with negative returns for two consecutive years have a higher risk of shutting down. Liang (2000) finds that the annual hedge-fund attrition rate is 8.3% for the 1994-1998 sample period using TASS data, and Baquero, Horst, and Verbeek (2002) find a slightly higher rate of 8.6% for the 1994-2000 sample period. Baquero, Horst, and Verbeek (2002) also find that surviving funds outperform non-surviving funds by approximately 2.1% per year, which is similar to the findings of Hsieh (2000, 2002b) and Liang (2000), and that investment style, size, and past performance are significant factors in explaining survival rates.…”