¹School of Information Science and Engineering, Shandong Normal University, Jinan 250014, China.
²Shandong Provincial Key Laboratory for Novel Distributed Computer Software Technology, Jinan 250014, China.
³Institute of Data Science and Technology, Shandong Normal University, Jinan 250014, China.

---

Abstract—The fluctuations in prices in financial derivative market is an important indicator to a country or region’s economic development, hence, predication of prices of financial derivatives is the research focus at present. However, the fluctuations in prices in financial market has high degree of nonlinearity, the traditional mathematical model has limitation to some extent, which influences the accuracy of prediction. This paper uses characterized Candlestick technique to implement noise removal processing on financial data and combines cooperative coevolution algorithms (CCEA) and support vector machine (SVM) to acquire the accuracy of classified prediction. After the noise removal processing made by characterized Candlestick technique, the core features of financial time series have been extracted, the randomness of financial data has been reduced, and the complexity of modeling has been simplified. The combination of CCEA and SVM has lifted the parameter optimization performance, acquired higher accuracy of classified prediction, and fit the solution of complex models. According to the computer simulation experiment, real stock data is used to verify the algorithm above, which has proved the high accuracy of prediction of this algorithm and the universality of this model.

Index Terms—Classified prediction, support vector machine, characterized candlestick, cooperative coevolution algorithms.

[PDF]

Cite: Jiang Zhipeng, Luo Chao, "Financial Time Series Forecasting Based on Characterized Candlestick and the Support Vector Classification with Cooperative Coevolution," Journal of Computers vol. 14, no. 3, pp. 195-209, 2019.