A MULTISCALE MODELING APPROACH INCORPORATING ARIMA AND ANNS FOR FINANCIAL MARKET VOLATILITY FORECASTING

doi:10.1007/s11424-014-3305-4

Journal of Systems Science and Complexity ›› 2014, Vol. 27 ›› Issue (1): 225-236.DOI: 10.1007/s11424-014-3305-4

• article • Previous Articles

A MULTISCALE MODELING APPROACH INCORPORATING ARIMA AND ANNS FOR FINANCIAL MARKET VOLATILITY FORECASTING

XIAO Yi¹ , XIAO Jin² , LIU John³, WANG Shouyang⁴

1.School of Information Management, Central China Normal University, Wuhan 430079, China; 2.Business School, Sichuan University, Chengdu 610064, China; 3.Center for Transport Trade and Financial Studies, City University of Hong Kong, Hong Kong, China; 4. Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing 100190, China.

Received:2012-04-01 Online:2014-02-25 Published:2014-08-20
Supported by:
This research is supported by the Humanities and Social Sciences Youth Foundation of the Ministry of Education f PR of China under Grant No. 11YJC870028, the Selfdetermined Research Funds of CCNU from the Colleges’ Basic Research and Operation of MOE under Grant No. CCNU13F030, China Postdoctoral Science Foundation under Grant No. 2013M530753 and National Science Foundation of China under Grant No. 71390335.

PDF ( 452 ) Cited

XIAO Yi , XIAO Jin , LIU John , WANG Shouyang. A MULTISCALE MODELING APPROACH INCORPORATING ARIMA AND ANNS FOR FINANCIAL MARKET VOLATILITY FORECASTING[J]. Journal of Systems Science and Complexity, 2014, 27(1): 225-236.

The financial market volatility forecasting is regarded as a challenging task because of irregularity, high fluctuation, and noise. In this study, a multiscale ensemble forecasting model is proposed. The original financial series are decomposed firstly different scale components (i.e., approximation and
details) using the maximum overlap discrete wavelet transform (MODWT). The approximation is predicted by a hybrid forecasting model that combines autoregressive integrated moving average (ARIMA) with feedforward neural network (FNN). ARIMA model is used to generate a linear forecast, and then
FNN is developed as a tool for nonlinear pattern recognition to correct the estimation error in ARIMA forecast. Moreover, details are predicted by Elman neural networks. Three weekly exchange rates data are collected to establish and validate the forecasting model. Empirical results demonstrate consistent better performance of the proposed approach.

No related articles found!

Viewed

Full text

Abstract