22 Pages Posted: 23 Apr 1998
Date Written: January 9, 1997
We use an iterative relocation algorithm to identify factors in common stock returns. The benefit of the approach is that factors are portfolios of assets with non-negative weights. As a result, they are readily interpretable in terms of the characteristics of the underlying securities. The positive portfolio factors have comparatively high explanatory power in sample and out of sample. We find evidence of a size factor and factors identified with certain industries. Factors extracted from the mutual fund universe perform marginally better than factors from the universe of equities.
JEL Classification: G1
Suggested Citation: Suggested Citation
Brown, Stephen J. and Goetzmann, William N. and Grinblatt, Mark, Positive Portfolio Factors (January 9, 1997). Yale School of Management Working Paper No. F-57. Available at SSRN: https://ssrn.com/abstract=79373 or http://dx.doi.org/10.2139/ssrn.79373