Designing a hybrid model for stock price prediction based on time series methods and deep artificial neural network in Tehran Stock Exchange

Document Type : Original Article

Authors
jina ahmadi 1
Hasan ghodrati ghezaani 2
Mehdi madanchi zaj 3
hossein jabbari 4
Aliakbar farzinfar 4
1 PhD Candidate in Financial Engineering, Department of Management, Kashan Branch, Islamic Azad University, Kashan, Iran
2 Hassan Ghodrati Ghaza’ani Assistant Professor, Department of Management, Kashan Branch, Islamic Azad University, Kashan, Iran
3 Assistant Professor, Department of Financial Management, Central Tehran Branch, Islamic Azad University, Tehran, Iran
4 Assistant Professor, Department of Accounting and Management, Kashan Branch, Islamic Azad University, Kashan, Iran
10.30495/jik.2024.69744.4409
Abstract
The purpose of this article is to predict stock prices based on time series and neural network methods in Tehran Stock Exchange. Stock price is considered as the main factor in investment decisions in the stock market.Therefore, it is very important to identify the factors affecting stock prices. The stock price shows the capital market's ability to absorb the available liquidity in the society and is often a symbol of the market's expectations about the economic status of the companies.Also, the stock price and its changes are an expression of market risk. Accordingly, in this study, two methods of Markov regime change and pseudo-neural with deep learning were used in the period of 1390-1400 in order to predict stock prices.The statistical population of the present study was the companies admitted to the Tehran Stock Exchange. The statistical sample includes 115 companies selected in the mentioned time period. The results obtained from the estimated model indicated that the designed hybrid model had a higher predictive power than the time series and neural network models.
Keywords
Stock price
prediction
regression
1.      آذر، عادل و افسر، امیر (1385)، مدل‌سازی پیش‌بینی قیمت سهام با رویکرد شبکه‌های عصبی فازی، فصلنامه پژوهش‌نامه بازرگانی، 40، 33-52.
2.      باباجانی، جعفر، تقوا، محمدرضا، بولو، قاسم، عبدالهی، محسن (1398)، پیش بینی قیمت سهام در بورس تهران با استفاده از شبکه عصبی بازگشتی بهینه شده با الگوریتم کلونی زنبور عسل مصنوعی. راهبرد مدیریت مالی، 7(2)، 195-228.
3.       دولو، مریم و حیدری، تکتم (1396)، پیش‌بینی شاخص سهام با استفاده از ترکیب شبکه عصبی مصنوعی و مدل‌های فرا ابتکاری جستجوی هارمونی و الگوریتم ژنتیک، اقتصاد مالی، 11(40)، 1-24.
4.      دولو، مریم، صفری، علی (1395)، پیش بینی شاخص قیمت سهام با استفاده از مدل هیبریدی. فصلنامه بورس اوراق بهادار، 9(35)، 81-101.
5. عباس­پور، محمد رضا (1381)، پیش بینی قیمت سهام شرکت ایران خودرو به کمک شبکه های عصبی، پایان نامه کارشناسی ارشد اقتصاد، دانشگاه تربیت مدرس.
6.       مهرآرا، محسن، معینی، علی، احراری، مهدی (1388)، الگوسازی و پیش بینی شاخص بورس اوراق بهادار تهران و تعیین متغیرهای موثر بر آن، پژوهش­ها و سیاستهای اقتصادی، 17(50)، 51-31.
7.      مهربان پور، محمدرضا، آذر، عادل، شهرامی بابکان، مجید (1401)، پیش‌بینی قیمت سهام با ارائه مدلی ترکیبی با استفاده از تحلیل مؤلفه‌های اصلی و تئوری مجموعه‌های راف، پژوهش های نوین در تصمیم گیری، 7(2)، 137-167.
8.      وطن پرست، محمدرضا، اسدی، مسعود، محمدی، شعبان، بابایی، عباس (1398)، پیش بینی قیمت سهام بر اساس شبکه عصبی LM-BP و برآورد نقطه بیش از حد توسط شمارش فواصل زمانی: شواهدی از بورس اوراق بهادار، مهندسی مالی و مدیریت اوراق بهادار، 10(39)، 193-218.
9.      هال، جان (1384)، مبانی مهندسی مالی و مدیریت ریسک، ترجمه: سجاد سیاح و علی صالح­آبادی، بخش تحقیقات شرکت کارگزاری مفید، تهران: گروه رایانه تدبیرپرداز.
10.  Abbasi, E., & Abouec, A. (2018), "Stock price forecast by using neuro-fuzzy inference system", Proceedings of World Academy of Science, Engineering and Technology, No. 36, PP 320–323.
11.  Atsalakis, G. S., & Valavanis, K. P. (2019), "Forecasting stock market short term trends using a neuro-fuzzy based methodology", Expert Systems with Applications, No. 36(3), PP 10696–10707.
12.  Bley, J., (2019), "Are GCC Stock Markets Predictable?", Emerging Markets Review, Vol. 12, PP. 217–237.
13.  Boyacioglu, M.C., Avci, D. (2010), "An Adaptive Network-Based Fuzzy Inference System (ANFIS) for the prediction of stock market return: The case of the Istanbul Stock Exchange", Expert Systems with Applications, No. 37, pp 7908–7912.
14.  Briatka, L., (2006), "How Big is Big Enough? Justifying Results of the i.i.d Test Based on the Correlation Integral in the Non-Normal World", Working Paper Series, ISSN. 1211-3298, No. 308, PP. 1-34.
15.  Cadenas, Erasmo & Rivera, Wilfrido (2010),” Wind speed forecasting in three different regions of Mexico, using a hybrid ARIMA-ANN model”, Renewable Energy, No. 35.
16.  Chuang, W.I., Liu, H.H., Susmel, R., (2012), "The bivariate GARCH approach to investigating the relation between stock returns, trading volume, and return volatility", Global Finance Journal, In Press, Uncorrected Proof, Available online 13, Elsevier.
17.  Guresen, Erkam & Kayakutlu, Gulgun & Daim, u. Tugrul (2021),” Using artificial neural network models in stock market index prediction”, Expert Systems with Applications.
18.  Jibendu Kumar Mantri, Dr. P.Gahan and B.B.Nayak, (2010),"Artificial neural networks- an application to stock market volatility", International Journal of Engineering Science and Technology, Vol. 2(5), pp 1451-1460.
19.  Kumar, Ujjwal & Jain, V. K (2010),” ARIMA forecasting of ambient air pollutants (O3, NO, NO2 and CO)”, publisher: Springer.
20.  Lee, J.W., Lee, K.E., Rikvold, P.A., (2004), "Multifractal Behavior of the Korean Stock market Index KOSPI", Working Paper.
21.  Qui-yong Z., Xiaoyu Z., Duan F.(2015), "Prediction Model of Stock Prices Based on Correlative Analysis and Neural Networks", Second International Conference on Information and Computing Science , pp: 189-192 , IEEE.
22.  Rubio, O.B., Rivero, S.S., Rodriguez, F.F., (2012), “Non-Linear Forecasting  Methods: Some Applications to the Analysis of Financial Series”, Fundacion de Estudios Economia Aplicada (FEDEA), Vol. 01, PP. 1-23.
23.  Tong-Seng Q. (2015), "Using Neural Network for DJIA Stock Selection ", Engineering Letters, EL_15_1_19, pp 15:31.
24.  Xiu, J., Jin, Y., (2017), "Empirical Study of ARFIMA Model Based On Fractional Differencing", Physica-A, No.  377, PP. 137-184.
25.  Yamashita T., Kotaro Hirasawa, Jinglu Hu (2017), "Application of Multi-Branch Artificial neural networksto Stock Market Prediction", Proceedings of International Joint Conference on Neural Networks, Montreal, Canada.
26.  Ying Wei, L., Liang Chen, T., Hwa Ho, T. (2020), "A hybrid model based on adaptive-network-based fuzzy inference system to forecast Taiwan stock market", Expert Systems with Applications, No. 38, pp 13625–13631.
27.  Zhou, Jian & Kang, Zhixin (2020),” A Comparison of Alternative Forecast Models of REIT Volatility”, Journal of Real Estate Finance Economics, PP. 275-294.