Quarterly Journal of New Approaches
in Industrial Engineering and Management
Volume 3, Issue 4 (3-2026)
2026, 3(4): 50-59 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Qanbarzadeh Anbarani E. Hedging with linear regressions and neural networks. Quarterly Journal of New Approaches in Industrial Engineering and Management 2026; 3 (4) :50-59
URL: http://iem-science.ir/article-1-224-en.html
URL: http://iem-science.ir/article-1-224-en.html
Master of Financial Mathematics, Tarbiat Modares University, Tehran, Iran
Abstract: (219 Views)
In this study, neural networks are investigated as nonparametric estimation tools for hedging options. For this purpose, a network called HEDGENET is designed, the output of which is a hedging strategy. The network is trained to minimize the hedging error rather than the pricing error. Applied to the time horizon and spot prices of the Euro Stoxx 50 and S&P 500 option indices, the network is able to significantly reduce the mean squared error of the Black-Scholes hedge. However, a similar advantage can be achieved through simple linear regressions accounting for the effect of leverage
Keywords: Modeling, Black-Scholes, Hedging error, Information leakage, Leverage effect, Statistical hedging
Type of Study: Research |
Subject:
Special
Received: 2026/01/17 | Accepted: 2026/03/16 | Published: 2026/03/18
Received: 2026/01/17 | Accepted: 2026/03/16 | Published: 2026/03/18
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |