Speaker: Dr Soohun Kim (KAIST College of Business)
Hosted by: University of Liverpool Management School's Accounting and Finance Group
Open to: Management School PhD students and academic staff, with no sign up needed
Date: Wednesday 14 February 2024
Time: 3-4.30pm
Place: Management School - Seminar Room 5
Abstract
This paper offers a novel approach to identifying the relationship between extensive option panels and market returns using functional predictive regression.
Employing our approach on the options and realized returns of the S&P 500, we achieve remarkable performance in predicting S&P 500 monthly returns, yielding an in-sample R2 of 5.12\% and an out-of-sample $R^2$ of 4.88\%.
Additionally, our method proves highly effective in predicting the variance risk premium of the S&P 500 index, achieving in-sample and out-of-sample R2 values of 33.65\% and 14.29\%.
The predictive accuracy of our model surpasses that of established predictors and equilibrium models, providing substantial utility gains over benchmark approaches in out-of-sample contexts.
We find that both the use of options panels and the adoption of functional regression are indispensable for the outperformance.
Co-authored by Yoosoon Chang, Youngmin Choi, Soohun Kim and Joon Park.
Speaker
Soohun Kim is an associate professor of finance at the College of Business at KAIST. Prior to joining KAIST, he was on the faculty at the Scheller College of Business at Georgia Tech.
He is the Associate Editor for Asia-Pacific Journal of Financial Studies (AJFS) by Korean Securities Association and the Associate Editor for Asian Review of Financial Research (ARFR) by Korean Finance Association.
His current research interest are in the area of asset pricing, financial econometrics and machine learning, and tail risk in financial markets.
Back to: Management School