Talk:Survivorship bias

--ThePrune 10:28, 22 September 2011 (EDT) It would be nice to include one or two specific examples of how ignoring dead mutual funds or dead hedge funds distorts the calculation of anticipated average returns.


 * It appears that the findings for mutual funds are all over the map.


 * I am trying to absorb the findings reported in the Grubar, Elton, Blake paper but I am somewhat confused by the multiple methodologies examined. The period measured is 1976-1993. The paper has a section devoted to a review of prior studies (along with their methodology problems). Here are the numbers:


 * Brown & Goetzmann: examining raw returns fund data from 1976-1988 found bias of 20 and 80 basis points, depending on the weighing scheme. Malkiel (1994) using unadjusted raw data found bias of 150 basis points.


 * Grubar, Elton, Blake: Apparently modelers attempting to determine bias measure performance either by considering the common stock investing policy at the beginning of the period or by considering the common stock investing policy throughout the period. The study authors report both approaches. The authors prefer a risk-adjusted three index method (passive benchmarks for large cap stocks, small cap stocks, bonds and bills) over all other methods (such as single index, S&P 500 index, or raw returns) of measuring bias. Panel A looks at results assuming investment ends at the point the fund disappears from the data record via merger or policy change. Panel B shows a "follow the money" method, assuming the the investment continues to be held in the new merged fund after the disappearing fund has been merged. In the paper, GE&B report the survivorship bias they found in bond funds, using similar risk-adjusted three index factor methodology (in a 1993) paper as 27 basis points of raised return. Here is a reproduced table from the paper, showing results for the period employing various methodologies.



--Blbarnitz 12:45, 22 September 2011 (EDT)
 * + Alternate Measures of Survivorship Bias (1976-1993) from Grubar, Elton, and Blake
 * align="center" style="background:#f0f0f0;"|Panel A. Through month of merger or policy change
 * align="center" style="background:#f0f0f0;"|Common stock at beginning
 * align="center" style="background:#f0f0f0;"|Common stock throughout
 * 1-index alpha||0.3199||0.0274
 * 3-index alpha||0.9069||0.7336
 * Raw Return||1.8430||2.2817
 * Raw Return minus S&P||0.3908||-0.4190
 * Panel B. Follow the money approach||Common stock at beginning||Common stock throughout
 * 1-index alpha||0.7344||0.6802
 * 3-index alpha||0.7716||0.7055
 * Raw Return||0.7301||0.5982
 * Raw Return minus S&P||0.7301||0.5982
 * }
 * 1-index alpha||0.7344||0.6802
 * 3-index alpha||0.7716||0.7055
 * Raw Return||0.7301||0.5982
 * Raw Return minus S&P||0.7301||0.5982
 * }
 * Raw Return||0.7301||0.5982
 * Raw Return minus S&P||0.7301||0.5982
 * }
 * }

Consider the work by Morningstar, which was challenged by Savant Capital Management. See this thread: [http://www.bogleheads.org/forum/viewtopic.php?t=76056 How Much? - Survivorship Bias at Morningstar], which references Study: Hidden Bias in Morningstar Data ' Systematically and Significantly' Overstates Managed Mutual Fund Performance

Forum thread: Survivorship Bias contains tutorial examples. This post is looking for a wiki article.

--LadyGeek 21:36, 23 September 2011 (EDT)