Forecasting and analysis of the volatility of the Shanghai Composite Index by semi-parametric methods
Shi Wang · Journal of Beijing University of Chemical Technology · 2011
The errors in a financial index over time always exhibit heteroscedastic properties.In this paper,we discuss the volatility of the Shanghai Stock Exchange(SSE) Composite Index using parametric and semi-parametric methods,and compare the sample fitting and forecasting ability of the methods on the basis of four measurement indices for forecasting error with samples of the SSE Composite Index.The results showed that the semi-parametric model is superior to the parametric model,and that a semi-parametric additive model is superior to a polynomial spline model.