%0 Journal Article %A Xiaohui Ni %A Yannick Malevergne %A Didier Sornette %A Peter Woehrmann %T Robust Reverse Engineering of Cross-Sectional
Returns and Improved Portfolio Allocation
Performance Using the CAPM %D 2011 %R 10.3905/jpm.2011.37.4.076 %J The Journal of Portfolio Management %P 76-85 %V 37 %N 4 %X Explaining the cross-sectional characteristics of asset returns within the mean-variance framework is the holy grail of financial economists and investment managers. Empirical analysis, however, has shown that the CAPM does not work in practice. In 2010, Levy and Roll suggested that the inadequacy of the CAPM could just result from statistical errors; that is, due to the large statistical errors of the inputs of Markowitz’ portfolio optimization, its results are not reliable and should be considered very cautiously. This assertion led to turning the usual approach on its head, positing that the market portfolio is actually mean-variance efficient, and asking how much the inputs should be tinkered with to allow this. In this article, the authors present a series of tests and applications. They apply the Levy–Roll procedure to the 25 Fama–French portfolios sorted by size and book-to-market value. They check the consistency of the Levy–Roll approach by investigating how the adjusted stock returns of specific stocks are modified when varying the basket of stocks they belong to. They then test the dynamic performance of the Levy–Roll procedure over the period from January 1992 to December 2009.And, finally, they show how to exploit the method for better portfolio allocation.TOPICS: Analysis of individual factors/risk premia, factor-based models, statistical methods %U https://jpm.pm-research.com/content/iijpormgmt/37/4/76.full.pdf