Research Article | | Peer-Reviewed

Forecasting Stock Prices Using Heston-Artificial Neural Network Model

Received: 13 September 2023     Accepted: 4 October 2023     Published: 28 October 2023
Views:       Downloads:
Abstract

Considering the evolution of financial globalization and the impacts of the global economic crisis, stock trading faces unprecedented fluctuations. The inherent volatility in stock prices has resulted in market uncertainty, prompting an interest among investors in reliable pricing models in order to maximize profits. To this end, researchers have continued to diligently refine stock pricing models to mitigate market uncertainty. One notable contender in this arena is the Heston model, conceived to remedy the limitations of the Black-Scholes model. The model embraces stochastic volatility, a departure from the constant volatility assumption underpinning the Black-Scholes model. However, the Heston model itself grapples with certain pivotal constraints, mainly the requisite precision in parameter calibration to produce a reliable estimate. Leveraging the current wave of technological advancement, this study uses an Artificial Neural Network (ANN) as a substitute for simulating different volatility parameters in the Heston model. This approach culminates in the construction of a hybrid semi-parametric forecasting model termed the Heston-ANN model. The study applies this model to datasets of three distinct stocks: BA, IBM, and GOLD. Through graphical analysis and the evaluation of different model performance metrics including Mean Absolute Percentage Error, Mean Absolute Error, and Mean Squared Error, the study compares the hybrid model to the original Heston model. The results reveal that the Heston-ANN model yields more accurate forecasts when juxtaposed with its precursor, the original Heston model. The synergy between the Heston model and ANN makes the hybrid model a more robust solution for forecasting stock prices.

Published in International Journal of Data Science and Analysis (Volume 9, Issue 2)
DOI 10.11648/j.ijdsa.20230902.11
Page(s) 22-33
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2023. Published by Science Publishing Group

Keywords

Stock Pricing, Artificial Neural Network (ANN), Stochastic Volatility, Stochastic Differential Equation (SDE)

References
[1] I-Ming Jiang, Jui-Cheng Hung, and Chuan-San Wang. Volatility forecasts: Do volatility estimators and evaluation methods matter? Journal of Futures Markets, 34 (11): 1077–1094, 2014.
[2] Shunrong Shen, Haomiao Jiang, and Tongda Zhang. Stock market forecasting using machine learning algorithms. Department of Electrical Engineering, Stanford University, Stanford, CA, pages 1–5, 2012.
[3] Chandan Sengupta. Financial modeling using excel and VBA. John Wiley & Sons, 2004.
[4] G Preethi and B Santhi. Stock market forecasting techniques: A survey. Journal of Theoretical & Applied Information Technology, 46 (1), 2012.
[5] GS Ladde and Ling Wu. Development of modified geometric Brownian motion models by using stock price data and basic statistics. Nonlinear Analysis: Theory, Methods & Applications, 71 (12): e1203–e1208, 2009.
[6] Dean Teneng. Limitations of the Black-Scholes model. Collection of Papers, 1: 143, 2011.
[7] Beni Lauterbach and Paul Schultz. Pricing warrants: an empirical study of the Black-Scholes model and its alternatives. The Journal of Finance, 45 (4): 1181–1209, 1990.
[8] Steven L Heston. A closed-form solution for options with stochastic volatility, with applications to bond and currency options,” review of financial studies 6. 1993.
[9] Pierre Gauthier and Dylan Possama¨ı. Efficient simulation of the double Heston model. Available at SSRN 1434853, 2010.
[10] Fernando Ormonde Teixeira. On the numerical methods for the Heston model. PhD thesis, 2017.
[11] Naiga Babra Charlotte, Joseph Mung’atu, Nafiu Lukman Abiodun, and Mark Adjei. On modified Heston model for forecasting stock market prices. 2022.
[12] Bianca Reichert. Can the Heston model forecast energy generation?: a systematic literature review. International Journal of Energy Economics and Policy, 2022.
[13] Luca Di Persio and Oleksandr Honchar. Artificial neural networks approach to the forecast of stock market price movements. International Journal of Economics and Management Systems, 1, 2016.
[14] Jordan Ayala, Miguel Garc´ıa-Torres, Jos´e Luis V´azquez Noguera, Francisco G´omezVela, and Federico Divina. Technical analysis strategy optimization using a machine learning approach in stock market indices. Knowledge-Based Systems, 225: 107119, 2021.
[15] Roumen Trifonov, Radoslav Yoshinov, Galya Pavlova, and Georgi Tsochev. Artificial neural network intelligent method for prediction. In AIP Conference Proceedings, volume 1872, page 020021. AIP Publishing LLC, 2017.
[16] Elias M Stein and Jeremy C Stein. Stock price distributions with stochastic volatility: an analytic approach. The review of financial studies, 4 (4): 727–752, 1991.
[17] Brandon Hardin. Implementing the Heston option pricing model in object-oriented cython. 2017.
[18] CG Looney. Pattern recognition using neural networks: theory and algorithms for engineers and scientists: Oxford university press. New York, 1997.
[19] Kazuhisa Matsuda. Introduction to Merton jump diffusion model. Department of Economics. The Graduate Center, The City University of New York, 2004.
Cite This Article
  • APA Style

    Ann Maina, Samuel Mwalili, Bonface Malenje. (2023). Forecasting Stock Prices Using Heston-Artificial Neural Network Model . International Journal of Data Science and Analysis, 9(2), 22-33. https://doi.org/10.11648/j.ijdsa.20230902.11

    Copy | Download

    ACS Style

    Ann Maina; Samuel Mwalili; Bonface Malenje. Forecasting Stock Prices Using Heston-Artificial Neural Network Model . Int. J. Data Sci. Anal. 2023, 9(2), 22-33. doi: 10.11648/j.ijdsa.20230902.11

    Copy | Download

    AMA Style

    Ann Maina, Samuel Mwalili, Bonface Malenje. Forecasting Stock Prices Using Heston-Artificial Neural Network Model . Int J Data Sci Anal. 2023;9(2):22-33. doi: 10.11648/j.ijdsa.20230902.11

    Copy | Download

  • @article{10.11648/j.ijdsa.20230902.11,
      author = {Ann Maina and Samuel Mwalili and Bonface Malenje},
      title = {Forecasting Stock Prices Using Heston-Artificial Neural Network Model
    
    	
    },
      journal = {International Journal of Data Science and Analysis},
      volume = {9},
      number = {2},
      pages = {22-33},
      doi = {10.11648/j.ijdsa.20230902.11},
      url = {https://doi.org/10.11648/j.ijdsa.20230902.11},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijdsa.20230902.11},
      abstract = {Considering the evolution of financial globalization and the impacts of the global economic crisis, stock trading faces unprecedented fluctuations. The inherent volatility in stock prices has resulted in market uncertainty, prompting an interest among investors in reliable pricing models in order to maximize profits. To this end, researchers have continued to diligently refine stock pricing models to mitigate market uncertainty. One notable contender in this arena is the Heston model, conceived to remedy the limitations of the Black-Scholes model. The model embraces stochastic volatility, a departure from the constant volatility assumption underpinning the Black-Scholes model. However, the Heston model itself grapples with certain pivotal constraints, mainly the requisite precision in parameter calibration to produce a reliable estimate. Leveraging the current wave of technological advancement, this study uses an Artificial Neural Network (ANN) as a substitute for simulating different volatility parameters in the Heston model. This approach culminates in the construction of a hybrid semi-parametric forecasting model termed the Heston-ANN model. The study applies this model to datasets of three distinct stocks: BA, IBM, and GOLD. Through graphical analysis and the evaluation of different model performance metrics including Mean Absolute Percentage Error, Mean Absolute Error, and Mean Squared Error, the study compares the hybrid model to the original Heston model. The results reveal that the Heston-ANN model yields more accurate forecasts when juxtaposed with its precursor, the original Heston model. The synergy between the Heston model and ANN makes the hybrid model a more robust solution for forecasting stock prices.
    },
     year = {2023}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - Forecasting Stock Prices Using Heston-Artificial Neural Network Model
    
    	
    
    AU  - Ann Maina
    AU  - Samuel Mwalili
    AU  - Bonface Malenje
    Y1  - 2023/10/28
    PY  - 2023
    N1  - https://doi.org/10.11648/j.ijdsa.20230902.11
    DO  - 10.11648/j.ijdsa.20230902.11
    T2  - International Journal of Data Science and Analysis
    JF  - International Journal of Data Science and Analysis
    JO  - International Journal of Data Science and Analysis
    SP  - 22
    EP  - 33
    PB  - Science Publishing Group
    SN  - 2575-1891
    UR  - https://doi.org/10.11648/j.ijdsa.20230902.11
    AB  - Considering the evolution of financial globalization and the impacts of the global economic crisis, stock trading faces unprecedented fluctuations. The inherent volatility in stock prices has resulted in market uncertainty, prompting an interest among investors in reliable pricing models in order to maximize profits. To this end, researchers have continued to diligently refine stock pricing models to mitigate market uncertainty. One notable contender in this arena is the Heston model, conceived to remedy the limitations of the Black-Scholes model. The model embraces stochastic volatility, a departure from the constant volatility assumption underpinning the Black-Scholes model. However, the Heston model itself grapples with certain pivotal constraints, mainly the requisite precision in parameter calibration to produce a reliable estimate. Leveraging the current wave of technological advancement, this study uses an Artificial Neural Network (ANN) as a substitute for simulating different volatility parameters in the Heston model. This approach culminates in the construction of a hybrid semi-parametric forecasting model termed the Heston-ANN model. The study applies this model to datasets of three distinct stocks: BA, IBM, and GOLD. Through graphical analysis and the evaluation of different model performance metrics including Mean Absolute Percentage Error, Mean Absolute Error, and Mean Squared Error, the study compares the hybrid model to the original Heston model. The results reveal that the Heston-ANN model yields more accurate forecasts when juxtaposed with its precursor, the original Heston model. The synergy between the Heston model and ANN makes the hybrid model a more robust solution for forecasting stock prices.
    
    VL  - 9
    IS  - 2
    ER  - 

    Copy | Download

Author Information
  • Department of Statistics and Actuarial Sciences, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya

  • Department of Statistics and Actuarial Sciences, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya

  • Department of Statistics and Actuarial Sciences, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya

  • Sections