# Regulated LSTM Artificial Neural Networks for Option Risks

^{*}

## Abstract

**:**

## 1. Introduction

- A popular deep learning tool called LSTM, which is frequently used to forecast values in time-series data, is adopted to predict SSE50 ETF option prices.
- Instead of TIs “farming”, key technical indicators are optimally identified using two statistical methods for financial options: F-Regression and Mutual Information Regression. This idea improves the situation in which technical indicators are often included without considering their weights of contributions to forecasted results.
- A novel regulated LSTM model is proposed for option risks that combines the two different approaches with decision rules to increase pricing accuracy.
- The proposed model models are tested using real data (SSE50 ETF options) to demonstrate that the regulated LSTM model outperforms others.

## 2. Models

#### 2.1. B-S Model

#### 2.1.1. Volatility

#### 2.1.2. Risk-Free Interest Rate

#### 2.2. LSTM Neural Network

_{t}. It will possibly be added to the cell state.

_{t}to [−1, 1]. Finally, the output of the model can be gained by multiplying the output obtained by sigmoid pair by pair.

_{t−1}) pass according to a ft value, which ranges between 0 and 1. The LSTM neural network has an LSTM layer. It affects not only the output layer but also the adjacent neurons in the same layer. Since the financial data are not independent, the current price will affect the future price, so the existence of the LSTM layer is necessary.

- Activation function

- ReLU (Rectified Linear Unit)

- Leaky ReLU

- ELU (Exponential Linear Unit)

#### 2.3. Regulated LSTM Simulation

#### 2.3.1. Technical Indicators

- Williams Percent Range (Williams% R)

- Money Flow Index (MFI)

- Price Rate of Change (ROC)

- Chaikin Money Flow (CMF)

- Chande Momentum Oscillator (CMO)

- Detrended Price Oscillator (DPO)

#### 2.3.2. Optimal Selections of Technical Indicators

#### 2.4. Risk Measurements for Options

- ME (Mean Error)

- RMSE(Root Mean Squared Error)

- MAE(Mean Absolute Error)

- MAPE (Mean Absolute Percentage Error)

## 3. Empirical Analysis

## 4. Conclusions

## Author Contributions

## Funding

## Acknowledgments

## Conflicts of Interest

## References

- Bachelier, L. Theorie de la speculation. Ann. Sci. De L’ecole Norm. Super. Ser.
**1900**, 3, 21. [Google Scholar] [CrossRef] - Black, F.; Scholes, M.S. The pricing of options and corporate liabilities. J. Political Econ.
**1973**, 81, 637. [Google Scholar] [CrossRef] [Green Version] - Benaroch, M.; Kauffman, R.J. Justifying electronic banking network expansion using real options analysis. MIS Q.
**2000**, 24, 197. [Google Scholar] [CrossRef] [Green Version] - Kou, S. A jump diffusion model for option pricing. Manag. Sci.
**2002**, 48, 1086. [Google Scholar] [CrossRef] [Green Version] - Carr, P.; Wu, L. Time-changed levy processes and option pricing. J. Financ. Econ.
**2003**, 71, 113. [Google Scholar] [CrossRef] [Green Version] - Yao, J.; Li, Y.; Tan, C.L. Option price forecasting using neural networks. Omega-Int. J. Manag. Sci.
**2000**, 28, 455. [Google Scholar] [CrossRef] - Lin, C.T.; Yeh, H.Y. The valuation of Taiwan stock index option price—Comparison of performances between Black-Scholes and neural network model. J. Stat. Manag. Syst.
**2004**, 8, 355. [Google Scholar] [CrossRef] - Tan, D. S&P 500 index option pricing based on the BP neural networks. Stat. Inf. Forum
**2008**, 23, 40. [Google Scholar] - Liu, D.; Huang, S. The performance of hybrid artificial neural network models for option pricing during financial crises. J. Data Sci.
**2016**, 14, 1. [Google Scholar] [CrossRef] - Wang, L.L.; Li, Y. Research on option pricing of shanghai stock exchange 50ETF based on neural network. Comput. Knowl. Technol.
**2020**, 16, 196. [Google Scholar] - Hochreiter, S.; Schmidhuber, J. Long short-term memory. Neural Comput.
**1997**, 9, 1735. [Google Scholar] [CrossRef] [PubMed] - Graves, A.; Schmidhuber, J. Framewise phoneme classification with bidirectional LSTM and other neural network architectures. Neural Netw.
**2005**, 18, 602. [Google Scholar] [CrossRef] [PubMed] - Xie, H.L.; You, T. Research on European Stock Index options pricing based on deep learning algorithm: Evidence from 50ETF options markets. Stat. Inf. Forum
**2018**, 33, 99. [Google Scholar] - Fischer, T.; Krauss, C. Deep learning with long short-term memory networks for financial market predictions. Eur. J. Oper. Res.
**2018**, 270, 654–669. [Google Scholar] [CrossRef] [Green Version] - Zhu, B.; Pan, L. Test and analysis of a stock price forecasting indicator. Econ. Manag.
**2003**, 2, 68–73. [Google Scholar] - Ko, K.C.; Lin, S.J.; Su, H.J.; Chang, H.H. Value investing and technical analysis in Taiwan stock market. Pac. -Basin Financ. J.
**2014**, 26, 14–36. [Google Scholar] [CrossRef] - Nelson, D.M.Q.; Pereira, A.C.M.; de Oliveira, R.A. Stock market’s price movement prediction with LSTM neural networks. In Proceedings of the 2017 International Joint Conference on Neural Networks (IJCNN), Anchorage, AK, USA, 14–19 May 2017; pp. 1419–1426. [Google Scholar]
- Mohanty, D.K.; Parida, A.K.; Khuntia, S.S. Financial market prediction under deep learning framework using auto encoder and kernel extreme learning machine. Appl. Soft Comput.
**2020**, 99, 106898. [Google Scholar] [CrossRef] - Yıldırım, D.C.; Toroslu, I.H.; Fiore, U. Forecasting directional movement of Forex data using LSTM with technical and macroeconomic indicators. Financ. Innov.
**2021**, 7, 1. [Google Scholar] [CrossRef] - Piravechsakul, P.; Kasetkasem, T.; Marukatat, S.; Kumazawa, I. Combining Technical Indicators and Deep Learning by using LSTM Stock Price Predictor. In Proceedings of the 18th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), Chiang Mai, Thailand, 9–22 May 2021. [Google Scholar]
- Chriss, N.A.; Kawaller, I. Black-Scholes and Beyond: Option Pricing Models, 1st ed.; McGraw-Hill: New York, NY, USA, 1996. [Google Scholar]
- Barndorff-Nielsen, O.E.; Stelzer, R. The Multivariate supOU Stochastic Volatility Model. Ole Eiler Barndorff-Nielsen, Robert Stelzer. Math. Financ.
**2013**, 23, 275–296. [Google Scholar] [CrossRef] [Green Version] - Chorro, C.; Guégan, D.; Ielpo, F. Option pricing for GARCH-type models with generalized hyperbolic innovations. Quant. Financ.
**2012**, 12, 1079–1094. [Google Scholar] [CrossRef] [Green Version] - Ramponi, A. Spread Option Pricing in Regime-Switching Jump Diffusion Models. Mathematics
**2022**, 10, 1574. [Google Scholar] [CrossRef] - Ma, C.; Yue, S.; Ren, Y. Pricing Vulnerable European Options under Lévy Process with Stochastic Volatility. Discret. Dyn. Nat. Soc.
**2018**, 3402703. [Google Scholar] [CrossRef] [Green Version] - Schoutens, W. Lévy Processes in Finance: Pricing Financial Derivatives, 1st ed.; Wiley: Hoboken, NJ, USA, 2003. [Google Scholar]
- Liu, D. Markov modulated jump-diffusions for currency options when regime switching risk is priced. Int. J. Financ. Eng.
**2019**, 6, 1950038. [Google Scholar] [CrossRef] - China Assets Management Co Ltd. China 50 ETF. Available online: https://fund.chinaamc.com/portal/cn/uploadFiles/50ETF.1253167410245.pdf (accessed on 20 April 2022).
- A List of Technical Indicators. Available online: https://www.tradingtechnologies.com/xtrader-help/x-study/technical-indicator-definitions/list-of-technical-indicators/ (accessed on 12 April 2021).
- Salkar, T.; Shinde, A.; Tamhankar, N.; Bhagat, N. Algorithmic Trading using Technical Indicators. In Proceedings of the 2021 International Conference on Communication Information and Computing Technology (ICCICT), Mumbai, India, 25–27 June 2021; Available online: https://ieeexplore-ieee-org.ez.xjtlu.edu.cn/stamp/stamp.jsp?tp=&arnumber=9510135 (accessed on 28 November 2021).
- Brownlee, J. Better Deep Learning: Train Faster, Reduce Overfitting, and Make Better Predictions; Machine Learning Mastery. 2018. Available online: https://machinelearningmastery.com/better-deep-learning/#packages (accessed on 21 April 2022).
- Chen, Z.; Yang, C.; Qiao, J. The optimal design and application of LSTM neural network based on the hybrid coding PSO algorithm. J. Supercomput.
**2022**, 78, 7227–7259. [Google Scholar] [CrossRef]

TIs | Score |
---|---|

ROC_56 | 3684.859 |

RSI_56 | 3434.026 |

CMO_56 | 3422.589 |

KST_56 | 2607.816 |

DX_56 | 2267.488 |

CCI_56 | 2159.798 |

KDJK_56 | 2151.685 |

RSV_56 | 2034.888 |

WR_56 | 2034.888 |

MFI_56 | 1888.237 |

TRIX_56 | 1715.286 |

HMA_0 | 752.5516 |

TIs | Score |
---|---|

ROC_56 | 0.217495 |

WR_56 | 0.205305 |

RSV_56 | 0.204443 |

KST_56 | 0.203183 |

RSI_56 | 0.202899 |

CMO_56 | 0.199681 |

CCI_56 | 0.196113 |

TRIX_56 | 0.182988 |

HMA_0 | 0.181470 |

DX_56 | 0.180711 |

KDJK_56 | 0.172671 |

MFI_56 | 0.167238 |

Code | Date | Open | High | Low | Close | K | Maturity |
---|---|---|---|---|---|---|---|

510050C1806M02650 | 2 January 2018 | 0.3332 | 0.3575 | 0.3332 | 0.3575 | 2.65 | 27 June 2018 |

510050C1806M02650 | 3 January 2018 | 0.3637 | 0.3788 | 0.3568 | 0.3573 | 2.65 | 27 June 2018 |

Code | Date | Open | High | Low | Close | K | Maturity |
---|---|---|---|---|---|---|---|

510050P1806M02650 | 2 January 2018 | 0.0355 | 0.0355 | 0.0187 | 0.0249 | 2.65 | 27 June 2018 |

510050P1806M02650 | 3 January 2018 | 0.024 | 0.0247 | 0.02 | 0.0215 | 2.65 | 27 June 2018 |

Call Options | |||||||
---|---|---|---|---|---|---|---|

B-S | LSTM | Regulated LSTM | |||||

Predict | Train | Predict | Test | Train | Predict | Test | |

ME | 0.012715 | 0.000651 | 0.000717 | 0.000681 | 0.000375 | 0.000377 | 0.000377 |

RMSE | 0.112762 | 0.025505 | 0.026768 | 0.026092 | 0.019355 | 0.019405 | 0.019407 |

MAE | 0.095152 | 0.019400 | 0.020273 | 0.019916 | 0.014932 | 0.014828 | 0.014917 |

MAPE | 0.554104 | 0.127533 | 0.131535 | 0.129728 | 0.096781 | 0.095356 | 0.096492 |

Put Options | |||||||
---|---|---|---|---|---|---|---|

B-S | LSTM | Regulated LSTM | |||||

Predict | Train | Predict | Test | Train | Predict | Test | |

ME | 0.004177 | 0.000713 | 0.000723 | 0.000714 | 0.000281 | 0.000309 | 0.000280 |

RMSE | 0.064628 | 0.026709 | 0.026897 | 0.026717 | 0.016751 | 0.017582 | 0.016747 |

MAE | 0.049610 | 0.019972 | 0.020071 | 0.020023 | 0.011727 | 0.012066 | 0.011716 |

MAPE | 0.372994 | 0.158781 | 0.155234 | 0.157046 | 0.098828 | 0.102002 | 0.099301 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Liu, D.; Wei, A.
Regulated LSTM Artificial Neural Networks for Option Risks. *FinTech* **2022**, *1*, 180-190.
https://doi.org/10.3390/fintech1020014

**AMA Style**

Liu D, Wei A.
Regulated LSTM Artificial Neural Networks for Option Risks. *FinTech*. 2022; 1(2):180-190.
https://doi.org/10.3390/fintech1020014

**Chicago/Turabian Style**

Liu, David, and An Wei.
2022. "Regulated LSTM Artificial Neural Networks for Option Risks" *FinTech* 1, no. 2: 180-190.
https://doi.org/10.3390/fintech1020014