Too Good to Be True? Fallacies in Evaluating Risk Factor Models

70 Pages Posted: 3 Jan 2018

See all articles by Nikolay Gospodinov

Nikolay Gospodinov

Federal Reserve Bank of Atlanta

Raymond Kan

University of Toronto - Rotman School of Management

Cesare Robot

University of Georgia

Date Written: 2017-11-01

Abstract

This paper is concerned with statistical inference and model evaluation in possibly misspecified and unidentified linear asset-pricing models estimated by maximum likelihood and one-step generalized method of moments. Strikingly, when spurious factors (that is, factors that are uncorrelated with the returns on the test assets) are present, the models exhibit perfect fit, as measured by the squared correlation between the model's fitted expected returns and the average realized returns. Furthermore, factors that are spurious are selected with high probability, while factors that are useful are driven out of the model. Although ignoring potential misspecification and lack of identification can be very problematic for models with macroeconomic factors, empirical specifications with traded factors (e.g., Fama and French, 1993, and Hou, Xue, and Zhang, 2015) do not suffer of the identification problems documented in this study.

Keywords: asset pricing, spurious risk factors, unidentified models, model misspecification, continuously updated GMM, maximum likelihood, goodness-of-fit, rank test

JEL Classification: C12, C13, G12

Suggested Citation

Gospodinov, Nikolay and Kan, Raymond and Robot, Cesare, Too Good to Be True? Fallacies in Evaluating Risk Factor Models (2017-11-01). FRB Atlanta Working Paper No. 2017-9. Available at SSRN: https://ssrn.com/abstract=3092150

Nikolay Gospodinov (Contact Author)

Federal Reserve Bank of Atlanta ( email )

Atlanta, GA 30309
United States

HOME PAGE: https://www.frbatlanta.org/research/economists/gospodinov-nikolay.aspx?panel=1

Raymond Kan

University of Toronto - Rotman School of Management ( email )

105 St. George Street
Toronto, Ontario M5S 3E6 M5S1S4
Canada
416-978-4291 (Phone)
416-971-3048 (Fax)

Cesare Robot

University of Georgia

Register to save articles to
your library

Register

Paper statistics

Downloads
102
Abstract Views
388
rank
258,408
PlumX Metrics