Modeling univariate volatility of stock returns using stochastic GARCH models:Evidence from 4-Asian markets

• Main Author: Islam, Mohd Aminul
• Format: Article
• Language: English
• Published: INSI Publications 2013
• Subjects: HG4501 Stocks, investment, speculation
• Online Access: http://irep.iium.edu.my/33419/1/AJBAS_Published_294-303.pdf; http://irep.iium.edu.my/33419/; http://www.ajbasweb.com/
• Institution: Universiti Islam Antarabangsa Malaysia

This paper applies the Generalized Autoregressive Conditional Heteroscedasticity (GARCH) models to the financial time series (stock index returns) in four Asian markets namely; Kuala Lumpur Composite Index (KLCI) of Malaysia, the Straits Times Index (STI) of Singapore, Nikkei Indices (N225) of Japan and the Hang Seng Index (HSI) of Hong Kong. We included six years of data covering the period from January 2007 to December 2012. This comprises daily observations of 1477 for KLCI, 1493 for STI, 1469 for N225 and 1481 for HSI excluding the public holidays. Closing values for stock indices are used. We included most commonly used variations of conditional volatility models with imposing names such as Generalized Autoregressive Conditional Heteroscedasticity best known as the GARCH (1, 1), GARCH-in-Mean, Thresh-hold GARCH (TGARCH), Exponential GARCH (EGARCH) and Power GARCH models. We aim to empirically examine the use of GARCH models in capturing the stylized facts such as volatility clustering and leverage effects commonly observed in high frequency financial time series data. We found strong empirical evidence of volatility clustering and leverage effects in the daily stock index returns of all four markets showing that the daily stock index returns can be characterized by the GARCH models. The results from GARCH-in-Mean model, we find evidence of positive relationship between the expected risk and expected return in all four markets which is often predicted in investment theory.