International Journal of Nonlinear Analysis and Applications

Steel price volatility forecasting; application of the artificial neural network approach and GARCH family models

Document Type : Research Paper

Authors

Department of Accounting, Faculty of Economics and Accounting, Islamic Azad University, South Tehran Branch, Tehran, Iran

Abstract

GARCH family models are the most widely-used methods for forecasting price volatility. Given that this approach usually has extremely high forecast errors, continuous studies have been conducted to improve forecast models using different techniques. In the present manuscript, we expanded the fields of expert systems, forecast, and modeling using an artificial neural network (ANN) and Generalized Autoregressive Conditional Heteroskedasticity (GARCH) method that created an ANN-GARCH model. The hybrid ANN-GARCH model was used to forecast steel price volatility, and its accuracy was evaluated based on mean absolute error (MAE) and mean square error (MSE) evaluation criteria. The results indicated a general improvement in forecasting using ANN-GARCH compared to the GARCH method alone. The results were realized using copper price returns, the dollar index, gold price returns, and oil price returns as inputs. We also discussed the research implications for this field in addition to practical applications. The research results indicated better performance of the hybrid ANN/GARCH/N model than other models. Furthermore, the neural-network-based hybrid models could better forecast prices than other time series models.

Keywords

[1] M.J. Bakhtiaran and M. Zolfaghari, Designing a model to predict the return of the global gold price (with an emphasis on the combined models of the convolutional neural network and Garch family models), J. Financ. Engin. Secur. Manag. 13 (2022), no. 50, 92–117.
[2] R. Engle, Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation, Econometrica: J. Econ. Soc.50 (1982), no. 4, 987–1007.
[3] Th. Kriechbaumer, A. Angus and D. Parsons, An improved wavelet–ARIMA approach for forecasting metal, Monica Rivas Casado 39 (2015), 32–41.
[4] W. Kristjanpoller and P.E. Hernandez, Volatility of main metals forecasted by a hybrid ANN-GARCH model with regressors, Expert Syst. Appl. 84 (2017), 290–300.
[5] W. Kristjanpoller and M.C. Minutolo, Gold price volatility: A forecasting approach using the artificial neural network–GARCH model, Expert Syst. Appl. 42 (2014), no. 20, 7245–7251.
[6] W.D. Kristjanpoller and M.C. Minutolo, Forecasting volatility of oil price using an artificial neural network-GARCH model, Expert Syst. Appl. 65 (2016), 233–241.
[7] W.D. Kristjanpoller and M.C. Minutolo, A hybrid volatility forecasting framework integrating GARCH, artificial neural network, technical analysis and principal components analysis, Expert Syst. Appl. 109 (2018), 1–11.
[8] S. Lahmiri, Modeling and predicting historical volatility in exchange rate markets, Phys. A: Statist. Mech. Appl. 471 (2017), 387–395.
[9] S. Lahmiri and M. Boukadoum, An ensemble system based on hybrid EGARCH-ANN with different distributional assumptions to predict S&P 500 intraday volatility, Fluct. Noise Lett. 14 (2015), no. 1, 1550001.
[10] X. Lu, D. Que and G. Cao, Volatility forecast based on the hybrid artificial neural network and GARCH-type models, Procedia Comput. Sci. 91 (2016), 1044–1049.
[11] M.R. Moghadam, M. Manjezi and A.H. Mehrdanesh, Evaluation of factors affecting the price of iron ore using artificial neural networks, Third Conf. Open Pit Mines of Iran, Kerman, 2014.
[12] A.S. Mohammad Sharifi, K. Khalili Damghani, F. Abdi and S. Sardar, Bitcoin price forecasting using a hybrid ARIMA model and deep learning, Ind. Manag. Stud. 19 (2022), no. 61, 125–146.
[13] A. R. Mohammadi, S. Soltani Mohammadi and H. Bakhshande Amnieh, Iron ore price prediction using time series model, Mining Deve. 44 (2012), 88–92.
[14] R. Najarzadeh, M. Zolfaghari and S. Gholami, Designing a model to forecast the Stock Exchange with an emphasis on hybrid neural network and long-term memory models), Invest. Knowledge Quart. 9 (2020), no. 34.
[15] D.B. Nelson, Conditional heteroskedasticity in asset returns: A new approach, Econometrica: J. Econ. Soc. 59 (1991), no. 2, 347–370.
[16] F. Sanchez Lasheras, F. Javier de Cos Juez, A. Suarez Sanchez and A. Krzemien, Pedro Riesgo Fernandez forecasting the COMEX copper spot price by means of neural networks and ARIMA models, Resources Policy 45 (2015), 37–43.
[17] A. Sharif Far, M. Khalili Iraq, I. Raisi Vanani and M.F. Fala, Evaluation and validation of optimal deep learning architecture in stock price prediction (LSTM long term memory algorithm approach), Financ. Engin. Secur. Manag. 12 (2022), no. 48, 348–370.
[18] E. Tully and B.M. Lucey, A power GARCH examination of the gold market, Res. Int. Bus. Finance 21 (2007) no. 2, 316–325.
[19] H. Yan, N. Jian and W. Liu, A hybrid deep learning approach by integrating LSTM-ANN network with GARCH model for copper price volatility prediction, Phys. A: Statist. Mech. Appl. 557 (2020), 124907.
[20] H. Young Kim and Ch. Hyun Won, Forecasting the volatility of stock price index: A hybrid model integrating LSTM with multiple GARCH-type models, Expert Syst. Appl. 3 (2018), 25–37.
[21] S. Zoghi, R. Raei and S. Falahpor, Presenting a market direction prediction model for gold coin trades in Iran’s Commodity Exchange market using Long Short-Term Memory (LSTM) algorithm, Financ. Engin. Portfolio Manag. 13 (2022), no. 53, 34–53.
[22] M. Zolfaghari, B. Sahabi, and M.J. Bakhtyaran, Designing a model for forecasting the stock exchange total index returns (Emphasizing on combined deep learning network models and GARCH family models), Financ. Engin. Portfolio Manag. 11 (2020), no. 42, 138–171.
International Journal of Nonlinear Analysis and Applications
Volume 15, Issue 5
May 2024
Pages 189-204
  • Receive Date: 22 January 2023
  • Revise Date: 18 March 2023
  • Accept Date: 13 May 2023