Volatility Forecasting during Stock Disaster in China A-share Stock Market

BAO, JINGYI (2018) Volatility Forecasting during Stock Disaster in China A-share Stock Market. [Dissertation (University of Nottingham only)]

[img] PDF - Registered users only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (636kB)

Abstract

This study explores the performance of volatility prediction models in Chinses stock market during stock market crash period. The index of interest is CSI 300 and the time interval employed ranges from 09/08/2007 to 19/12/2008 (2008 financial crisis) and from 12/06/2015 to 10/05/2016 (China stock market crash in 2015) respectively. Besides, the models tested in this paper is EWMA, GARCH (1, 1) and EGARCH (1, 1). Both GARCH family models are fitted with t distribution and GED. The forecasting horizon is one day ahead in both periods with daily data and Parkinson estimator is treated as proxy of realized volatility. The way to evaluate performance of each model is based on MSE, QLIKE and HMAE loss functions. The results prove that EWMA has the most robust ability of volatility prediction in both periods under all three loss functions. Moreover, EGARCH models is generally superior to simple GARCH model.

Item Type: Dissertation (University of Nottingham only)
Keywords: volatility prediction, Chinese stock market, stock disaster
Depositing User: Bao, Jingyi
Date Deposited: 11 Mar 2022 16:26
Last Modified: 11 Mar 2022 16:26
URI: https://eprints.nottingham.ac.uk/id/eprint/53692

Actions (Archive Staff Only)

Edit View Edit View