Asymmetry Effects in Chinese Stock Markets Volatility: A Generalized Additive Nonparametric Approach

Midwest Finance Association Conference, USA, February 2010

35 Pages Posted: 24 Nov 2009 Last revised: 16 Mar 2010

Date Written: November 19, 2007

Abstract

The unique characteristics of Chinese stock markets give rise to the difficulty of assuming innovation distributions and the specification form of the volatility process when modelling volatility with the parametric GARCH family models. This paper examines the Chinese stock market volatility and the asymmetric effects in the Chinese stock volatility by a generalized additive nonparametric (GAM NP) smoothing technique. The empirical results indicates that the asymmetric effect exists in the Chinese stock market. However, the News Impact Curve from the GAM NP model indicates that limited amounts of good news are needed in order to to keep the market calm. Further, compared with other parametric and nonparametricmodels, the empirical result from the GAMNP model demonstrates better performance for the volatility forecasts, particularly in the out-of-sample forecast. This GAM NP technique should have wide applications to other merging stock markets which are similarly imperfect and incomplete.

Keywords: GARCH modeling, Nonparametric GARCH modeling, Generalized Additive

JEL Classification: C13, C14, G17, G19

Suggested Citation

Hou, Ai Jun, Asymmetry Effects in Chinese Stock Markets Volatility: A Generalized Additive Nonparametric Approach (November 19, 2007). Midwest Finance Association Conference, USA, February 2010. Available at SSRN: https://ssrn.com/abstract=1509394 or http://dx.doi.org/10.2139/ssrn.1509394

Ai Jun Hou (Contact Author)

Stockholm University ( email )

Universitetsvägen 10
Stockholm, Stockholm SE-106 91
Sweden

Register to save articles to
your library

Register

Paper statistics

Downloads
122
Abstract Views
963
rank
233,911
PlumX Metrics