Forecasting Model of Multivariate Chaotic Financial Time Series Based on DFNN
Aiming at the chaotic character of financial time series which make stock market forecasting a complex problem, a forecasting model of Shanghai composite index based on dynamic fuzzy neural network (OFNN) is proposed. R/S analysis is applied to study the Shanghai stock market. It is found that the price and volume of the stock index return show persistent property. The price and volumes delay time and embedding dimensions were determined by C-C method respectively and the phase space of multivariate financial time series was reconstructed to restore financial complex system. The input dimension of DFNN was decided by embedding dimension of multivariate financial stock index time series. It can improve precision and stability of prediction to use chaotic characteristic to deal with samples and mapping nonlinear function by DFNN which could achieve online learning, parameter estimation and structure identification. Fuzzy rules of financial system were determined by DFNN, which can reveal the financial system operating mode. Therefore, this research proves the effectiveness of the proposed model in the practical identification of financial chaotic system.
Chaotic financial time series forecasting Phase space econstruction DFNN Financial chaotic system identification
Bin Sun Tieke Li Yanyan Zhao Chunsheng Zhang
School of Economics and Management, University of Science and Technology Beijing, Beijing 100083, China Engineering Research Center of MES Technology for Iron & Steel Production, Ministry of Education, China
国际会议
成都
英文
1541-1545
2010-12-17(万方平台首次上网日期,不代表论文的发表时间)