69 Pages Posted: 20 Mar 2017 Last revised: 8 Sep 2017
Date Written: August 31, 2017
The asset pricing literature has produced hundreds of potential risk factors. Organizing this “zoo of factors" and distinguishing between useful, useless, and redundant factors require econometric techniques that can deal with the curse of dimensionality. We propose a model-selection method to systematically evaluate the contribution to asset pricing of any new factor, above and beyond what a high-dimensional set of existing factors explains. Our methodology explicitly accounts for potential model-selection mistakes, unlike the standard approaches that assume perfect variable selection, which rarely occurs in finite sample and produces a bias due to the omitted variables. We apply our procedure to a set of factors recently discovered in the literature. We show that several factors - such as profitability and investments - have statistically significant explanatory power beyond the hundreds of factors proposed in the past. In addition, we show that our risk price estimates and their significance are stable, whereas the model selected by simple LASSO is not.
Keywords: Factors, Risk Price, Post-Selection Inference, Regularized Two-Pass Estimation, Machine Learning, LASSO, Elastic Net
Suggested Citation: Suggested Citation
Feng, Guanhao and Giglio, Stefano and Xiu, Dacheng, Taming the Factor Zoo (August 31, 2017). Fama-Miller Working Paper Forthcoming; Chicago Booth Research Paper No. 17-04. Available at SSRN: https://ssrn.com/abstract=2934020 or http://dx.doi.org/10.2139/ssrn.2934020
By Andrew Ang