1. UKM Learning and Research Repository
2. PHD Thesis / Tesis PHD
3. Faculty of Science and Technology / Fakulti Sains dan Teknologi

Title:	Asymmetry, Heavy-Tailedness, And Structural Breaks In Garch Class Of Volatility Models: An Application In GCC Stock Markets
Authors:	Ajab Abdullah Alfreedi (P48714)
Supervisor:	Zaidi Isa, Professor Dr.
Keywords:	GARCH model Garch Class Of Volatility Models Asymmetry In Garch Class Of Volatility Models Heavy-Tailedness In Garch Class Of Volatility Models Structural Breaks In Garch Class Of Volatility Models Stock exchanges
Issue Date:	27-May-2013
Description:	This study generally examines stock market volatilities in GCC (Gulf Cooperation Council) economies. A statistical issue in the context of modeling stock market volatilities has been investigated in the literature over the past few decades due to its importance in stock market policies. Nevertheless, the existence of numerous theoretical and empirical studies in this context, however, until recently GCC countries seem to have been paid relatively little attention. Hence, this study attempts to re-address that imbalance with help of advances in the area of contemporary time series analysis. Here, it is important to note that a number of less attended research questions can be raised in this regard. The first is whether an asymmetry in financial markets has been affected by sudden changes in unconditional variance in GARCH estimations. Second is whether the volatility persistence in univariate GARCH models reduce under different conditional distribution assumptions. Third is what happens in case structural breaks are ignored in univariate symmetric, as well as, asymmetric GARCH estimations. Fourth is whether the stock market linkages exist among selected GCC countries. This investigation generally differs from the earlier studies in two ways. First, a methodological approach used in this study relies on the relatively recent advances in the field of applied time series analysis. Second, despite the fact of necessity, a limited number of studies seem to have investigated the topical issue for GCC countries. Specifically, this study uses two methodological approaches: univariate symmetric as well as asymmetric versions of GARCH class of models and multivariate symmetric as will as asymmetric GARCH-BEKK parametrization, with strong consideration on heavy-tailed distribution in both approaches. Turning to the research findings, several points should be noted. We have discovered that the persistence of volatility have decreased significantly in all GCC stock market volatility models when the regime changes are taken into account. Moreover, the estimation results generally reveal that the heavy-tailed conditional densities are found to be favored compared to the normal distribution. When the heavy-tailed distributions are assumed, the persistence has found to be reduced considerably in four countries' models out of total six cases in comparison to models under normal distribution. Hence, the distribution assumption also seems to play an important role in the estimation of persistence together with accounting for the regime shifts. Besides, for the analysis of asymmetry, this study relies on asymmetric parameters, as well as, news impact curves. The results suggest that both ways of analysis do not appear to contradict each other. This study also analyzes the market linkages impacts among the selected GCC countries. Findings suggest that an asymmetric model under multivariate heavy-tailed distribution behaves well in comparison to the rest of multivariate GARCH models considered in this study.,PhD
Pages:	252
Call Number:	HG4551 .A434 2013
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Science and Technology / Fakulti Sains dan Teknologi

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.