2021 Fiscal Year Final Research Report
Artifical Neural Network for Option Pricing
| Project/Area Number |
20K22138
|
|---|
| Research Category |
Grant-in-Aid for Research Activity Start-up
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
0107:Economics, business administration, and related fields
|
|---|
| Research Institution | Kanagawa University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2020-09-11 – 2022-03-31
|
| Keywords | 人口回路網 / デリバティブ / モンテカルロ法 / 漸近展開 / 機械学習 / 確率ボラティリティモデル / ディープラーニング |
|---|
| Outline of Final Research Achievements |
In this research, we propose a mixed approach of asymptotic expansion (AE) and artificial neural network (ANN) methods for derivative pricing to improve computational speed, stability, and approximation accuracy. We apply our method to European and barrier options when the underlying asset price follows stochastic volatility model. According to the results, our ANN enables more accurate predictions for derivatives prices and training becomes more robust and requires much less training data, only one in a hundred or one in a thousand, with smaller hyperparameters (i.e., layers and nodes), half or less, than previously proposed methods. We further examine our ANN method in conjunction with the SABR model, a popular stochastic volatility model that is widely used in financial practice, then present the simulation and experimental results to demonstrate the effectiveness of our approach.
|
| Free Research Field |
金融工学
|
| Academic Significance and Societal Importance of the Research Achievements |
金融実務におけるデリバティブの価格付けには，モンテカルロ法に代表される数値計算法や漸近展開などの近似手法が広く用いられている．しかし，前者は膨大な計算時間を要し，後者は満期が長く，ボラティリティが大きい商品では近似精度が著しく劣化するという難点がある．本手法を用いることで，従来は数値計算法に頼るしかなかった金融商品でも，近似解と同程度の計算速度で，安定的にデリバティブの価格を計算できるようになった．金融機関で扱われる金融商品は取引額が大きく，僅かな推定誤差でも巨額の損失が生じる可能性があることから，機械学習を精度良く高速に実行する必要があり，本手法の金融実務に対する貢献は大きい．
|
