Risks Factors that Matter: Textual Analysis of Risk Disclosures for the Cross-Section of Returns

Alejandro Lopez-Lira, Finance, The Wharton School

Abstract: I introduce risk factors that not only explain the cross-section of returns, but also unambiguously represent economic risk for the firms, are interpretable, and come directly from the companies. I exploit machine learning and natural language processing techniques to identify from the 10-K annual reports the types of risks that firms face, and to quantify how much each firm is exposed to each type of risk. I employ the exposure of companies to each type of risk to construct portfolios exposed to specific risks and construct a factor model with these portfolios. The model is not rejected by the set of 49 Industry Portfolios using the GRS test, which is not the case for the Five Factor Model of Fama and French (2015).

Read the full working paper here (PDF).

Michelle Eckert is Marketing and Communications Coordinator for the Mack Institute, where she works to engage students, researchers, and corporate partners in opportunities for collaboration. Michelle received her B.A. in Art from Valparaiso University in 2007. Her background includes two AmeriCorps terms of service working to teach mathematics, computer literacy, and job readiness skills to out-of-school youth in Philadelphia, focusing particularly on promoting access to post-secondary education.