Risk Factor Centrality and the Cross-Section of Expected Returns
51 Pages Posted: 2 Nov 2020 Last revised: 21 Jan 2021
Date Written: January 20, 2021
The Factor Zoo phenomenon calls for answers as to which risk factors are in fact capable of providing independent information on the cross-section of expected excess returns, while considering that asset-pricing literature has produced hundreds of candidates. Supported by the network analysis framework, our paper proposes a methodology that enables to select the elements that best summarize the information from the risk factor covariance matrix and investigate whether it is capable of explaining cross-section returns. Our findings lead to sparser models when compared to other factor zoo-related methodologies and, additionally, present equivalent in and out-of-sample R^2 results when confronted with those factor zoo models.
Keywords: Risk factors, factor zoo, graph lasso, network analysis
JEL Classification: G12, C55, D85
Suggested Citation: Suggested Citation