SSE Composite Index Prediction and Simulation System
Stock index forecasting is an important task in economic field; it is difficult to accurately predict index trends using the traditional prediction methods which are based on price and the quality. Studies on this field were described in this paper: histo
- PDF / 588,542 Bytes
- 12 Pages / 439.37 x 666.142 pts Page_size
- 56 Downloads / 189 Views
Abstract Stock index forecasting is an important task in economic field; it is difficult to accurately predict index trends using the traditional prediction methods which are based on price and the quality. Studies on this field were described in this paper: historical data and the influence of macroeconomic factors to stock price, fits the mathematical models of the principal component of the multi-lag regression and Sequence Cointegration and stock index predicts; At the same time the author has done the system simulation, results show that the mathematical model is very efficient and practical.
Keywords Multi-lag regression Timing cointegration Stock index forecasting Simulation system
1 Introduction The stock market has 100 years history from appearance to now in developed country, its development has been quite mature, while Chinese stocks listed in Shanghai Stock Exchange on December 19, 1990 issue. It is difficult and important to do stock price forecasting. Now the stock price forecasting theory and its research methods are mainly based on price and quantity. Through the study of the history of the stock price to forecast the future price trend based on the price, mainly has the following three kinds of methods: technical analysis, fundamental analysis of stock method, and artificial intelligence analysis; through the analysis of the influence of the price trend of macroeconomic factors based on quantitative An erratum to this chapter is available at 10.1007/978-3-642-37832-4_65 D. Huang (&) J. Jiang G. Wang College of Science, Naval University of Engineering, Wuhan 430033, China e-mail: [email protected]
F. Sun et al. (eds.), Knowledge Engineering and Management, Advances in Intelligent Systems and Computing 214, DOI: 10.1007/978-3-642-37832-4_64, Springer-Verlag Berlin Heidelberg 2014
709
710
D. Huang et al.
research, use mathematical method to select the influential variables, determine the corresponding the regression equation to imitate and forecast, mainly has the following three kinds of methods: time series forecasting method, cointegration analysis, and variable analysis method. Based on the analysis of price is mainly qualitative analysis, the analysis result is weak, lack of theoretical support, and with strong subjectivity. Based on the studies of volume, with limitations, sequential regression model assuming the time series pattern in the future and the past of the model is consistent with the hypothesis, in the long-term prediction is not practical. Multivariate analysis only considered and stock price index is related to several variables, as well as some of the information is not considered, along with the macroeconomic development and change, and stock index related variables are also changing. Cointegration analysis does not take consideration of time factor. Based on the above shortcomings, this paper considers the SSE Composite Index historical data and the influence of stock price volatility of macroeconomic factors were established based on the principal componen
Data Loading...
