My research is to explain stock returns in a regression model. My dependent variable is raw returns and abnormal returns calculated from 3-factor model (Fama and French). Besides regressing against my main explanatory variables, should I include betas of MKT, SMB and HML from abnormal returns estimation into the regression with many explanatory variables for raw returns and abnormal returns?
My question is based on papers where researchers seem to do like that (links below). In the paper of Garel and Petit-Romec, there are betas of MKT, SMB and HML in panels of regression results, while in the paper of Bae et al. there is an indication that factor loadings are included in all regressions (even for raw returns). Should I include the factor loadings as explanatory variables in regressions and why?
- Garel, Alexandre and Petit-Romec, Arthur, Investor Rewards to Environmental Responsibility: Evidence from the COVID-19 Crisis (June 5, 2020). Available at SSRN: https://ssrn.com/abstract=3620109
- Bae, Kee-Hong and El Ghoul, Sadok and Gong, Jason and Guedhami, Omrane, Does CSR Matter in Times of Crisis? Evidence from the COVID-19 Pandemic (December 1, 2020). Journal of Corporate Finance, Forthcoming, Available at SSRN: https://ssrn.com/abstract=3726390