MODELLING OF THE STOCK MARKET VOLATILITY RETURN IN THE PRESENCE OF OUTLIERS AND STRUCTURAL BREAKS: THE INDICATOR SATURATION APPROACH

MOHD NASIR, IDA NORMAYA (2020) MODELLING OF THE STOCK MARKET VOLATILITY RETURN IN THE PRESENCE OF OUTLIERS AND STRUCTURAL BREAKS: THE INDICATOR SATURATION APPROACH. Doctoral thesis, Univesiti Sains Malaysia.

Text Thesis PhD USM_ida normaya mohd nasir.pdf Restricted to Registered users only Download (5MB)

Abstract

GARCH model is often preferred and widely used by financial modelling professionals to predict the volatility movement of financial instruments. Despite the widespread use of GARCH, the estimation and the accuracy of the model are often distorted by the presence of outliers and structural changes in the data series. To overcome this issue, we proposed the indicator saturation approach to jointly detect outliers and structural changes in the context of volatile data. To the best of our knowledge, limited research has been carried out in detecting outliers and structural break in volatile data such as financial time series by incorporating in the GARCH framework, therefore this study attempts to fill this gap. This study used the supersaturation indicator (SSI) approach on both simulation and empirical data. The SSIGARCH approach was applied to the return series of stock indices, where the impact of outliers and structural breaks were assessed and compared between the Shariahcompliant indices and its conventional counterparts. Results showed that SSI-GARCH provides a better estimation of GARCH model as evidenced by the information criteria. This study is expected to contribute to the development of new techniques to detect structural and outliers in the volatility of financial time series thus improving the estimation of GARCH parameters.

Item Type:	Thesis (Doctoral)
Subjects:	Social Sciences > Social Sciences (General)
Depositing User:	ENCIK SAIFUL FADZLY JAMALUDIN
Date Deposited:	15 Jul 2026 07:30
Last Modified:	15 Jul 2026 07:30
URI:	https://repositori.mohe.gov.my/id/eprint/340

Actions (login required)

View Item