Intelligent Hybrid Model Based on Japanese Candlestick Patterns and Fuzzy Logic for Stock Market Trend Forecasting

Sadeghi, Zahra; Zare, Hashem; Khodaparast Shirazi, Jalil

doi:10.22075/ijnaa.2025.39838.5582

Intelligent Hybrid Model Based on Japanese Candlestick Patterns and Fuzzy Logic for Stock Market Trend Forecasting

Articles in Press

Document Type : Research Paper

Authors

Zahra Sadeghi

Hashem Zare

Jalil Khodaparast Shirazi

Department of Economics, Shi.C., Islamic Azad University, Shiraz, Iran

10.22075/ijnaa.2025.39838.5582

Abstract

The stock market is one of the most challenging areas for investors to predict and make decisions due to its dynamic, complex, and nonlinear nature. The aim of this research is to present an intelligent hybrid model based on Japanese candlestick patterns and a fuzzy inference system to predict the trend of the stock market. In this model, the qualitative characteristics of the candlestick (including the ratio of the body and shadows and the pattern type) are combined with the quantitative indicators of trend, momentum, and trading volume change. For each pattern, a power function was designed and modeled in the form of fuzzy linguistic variables. The fuzzy system, with multivariate rules and the center of gravity defuzzification method, produces the final buy, hold, or sell signal in the daily timeframe. Real market data were used after cleaning and dividing them into training and testing sets, and the performance of the proposed model was compared with single-axis models. The results showed that the hybrid model performs better in all indicators, including prediction accuracy (78.5%), percentage of successful trades (71.3%), and total return (+18.2%), and offers a higher profit-loss ratio (1.65) and a lower maximum capital loss (7.8%) than traditional models. The sensitivity analysis of the parameters also confirmed the stability and stability of the model against minor changes. Overall, the proposed hybrid model can be an efficient tool for financial analysts to generate accurate and flexible signals in uncertain market conditions.

Keywords

Stock market

fuzzy logic

Japanese candlestick patterns

technical analysis

hybrid model

trend forecasting

momentum

optimization

[1] C. Aranha and H. Iba, The memetic tree‑based genetic algorithm and its application to portfolio optimization, Memetic Comput. 1 (2009), no. 2, 139–151.

[2] V.P. Balqis, S. Subiyanto, and S. Supian, Optimizing Stock Portfolio with Markowitz Method as a Reference for Investment Community Decisions, Int. J. Res. Commun. Serv. 2 (2020), no. 2, 71–76.

[3] H. Chen, N. Jegadeesh, and R. Wermers, The value of active mutual fund management: An examination of the stockholdings and trades of fund managers, J. Financ. Quant. Anal. 35 (2021), no. 3, 343–368.

[4] T. Cura, Particle swarm optimization approach to portfolio optimization, Nonlinear Anal.: Real World Appl. 10 (2009), no. 4, 2396–2406.

[5] F.D. Freitasa, A.F. De Souzab, and A.R. de Almeida, Prediction‑based portfolio optimization model using neural networks, Lattice Comput. Natural Comput. 72 (2009), no. 10–12, 2155–2170.

[6] E. Kamo and C. Dagli, Fuzzy inference system for stock price prediction based on candlestick patterns, Int. J. Comput. Intell. Syst. 12 (2019), no. 1, 45–56.

[7] D. Kern and M. Müller, A continuous‑time portfolio optimization model under crash scenarios and parameter uncertainty: A worst‑case approach applied to the Nigerian stock market, J. Financ. Optim. 18 (2022), no. 3, 245–263.

[8] M. Koura, Solving the mean‑variance Markowitz portfolio model with integer constraints using particle swarm optimization, Comput. Econ. 59 (2020), no. 2, 321–338.

[9] B.S. Kumar and V. Ravi, A survey of the applications of text mining in financial domain, Knowledge‑Based Syst. 114 (2020), 128–147.

[10] Y. Lai, H. Chen, and L. Zhang, Solving the mean‑variance portfolio optimization problem with integer constraints using ant colony optimization, Expert Syst. Appl. 199 (2022), 116912.

[11] K.H. Lee and G.S. Jo, Expert system for predicting stock market timing using a candlestick chart, Expert Syst. Appl. 16 (2021), 357–364.

[12] T. Leung, K. Wong, and S. Chan, Improved user‑friendly estimations for Markowitz portfolio optimization and accurate return prediction: Evidence from the U.S. stock market, J. Invest. Res. 27 (2022), no. 4, 89–108.

[13] M.M. Madbouly, M. Elkholy, Y.M. Gharib, and S.M. Darwish, Predicting stock market trends for Japanese candlestick using cloud model, In: Hassanien, AE., Azar, A., Gaber, T., Oliva, D., Tolba, F. (eds) Proceedings of the International Conference on Artificial Intelligence and Computer Vision (AICV2020). AICV 2020. Advances in Intelligent Systems and Computing, vol 1153, Springer, Cham, 2020.

[14] M. Moradi, Portfolio Optimization in Tehran Stock Exchange by Water Cycle Algorithm, Financ. Manag. Perspect. 7 (2017), no. 20, 9–32.

[15] R. Naranjo, J. Arroyo, and M. Santos, Fuzzy modeling of stock trading with fuzzy candlesticks, Expert Syst. Appl. 82 (2017), 173–184.

[16] A. Pakmaram, S.J. Bahri, and M. Valizadeh, Selection and optimization of stock portfolio using genetic algorithm, using Markowitz mean‑half‑variance model, Financ. Engin. Secur. Manag. 8 (2017), no. 31, 19–42.

[17] M. Shahbazi, Stock Portfolio Optimization with Beta Coefficient Clustering Approach, Seventh National Conference on Modern Research in Management, Economics and Accounting, Iran, 2021.

[18] F. Ying, W. Yi‑Ming, and X. Wie‑Xuan, Application of VaR methodology to risk management in the stock market in China, Comput. Ind. Engin. 46 (2022), 383–388.

[19] X.S. Zhou and M. Dong, Can fuzzy logic make technical analysis 20/20?, Financ. Anal. J. 60 (2021), no. 4, 54–75.