One of our favorite websites, CXO Advisory, had a nice summary of O'Shaughnessy's Book.
From Chapter 21, “Using Multifactor Models to Improve Performance” (Page 470): “…you can do vastly better than a passive investment…by using more than one factor to select a portfolio of stocks. …Investors are best served by buying stocks that have jumped a series of hurdles rather than just one.”
http://www.cxoadvisory.com/17768/fundamental-valuation/a-few-notes-on-what-works-on-wall-street/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+cxo+%28CXO+Advisory+Blog+-+Investing+Notes%29
From Chapter 21, “Using Multifactor Models to Improve Performance” (Page 470): “…you can do vastly better than a passive investment…by using more than one factor to select a portfolio of stocks. …Investors are best served by buying stocks that have jumped a series of hurdles rather than just one.”
From Chapter 24, “Sector Analysis” (Page 545): “…what works in the All Stocks universe also works quite well at the sector level. …what we should avoid investing in at the All Stocks universe level should also be avoided at the sector level.”
From Chapter 25, “Searching for the Ideal Growth Strategy” (Page 567): “One of the very best ways to use price momentum is to marry it to a value constraint [composited value factors]. …six-month price appreciation is a more effective final momentum filter than 12-month price appreciation.”
While the author takes steps to mitigate data snooping bias, there are so many characteristics/combinations tested on the same data sets in search of best portfolio strategies that discrimination among strategies/variations may derive materially from luck.
No comments:
Post a Comment