STRUCTURAL BREAKS AND STOCK RETURN VOLATILITY IN MALAYSIA

Abstract

This study investigates the effect of structural breaks on volatility modelling in the Malaysia stock market using daily KLCI return data from the year 2001 to 2024. Knowing that the financial market often experiences sudden shifts due to economic crises, political transitions, and various global events, this study explores if such structural breaks can significantly influence the estimation of volatility under the ARCH framework. The Iterative Cumulative Sum of Squares (ICSS) algorithm is employed to detect multiple variance shifts in the KLCI return series before the ARCH-type models are estimated by incorporating the identified breakpoints. Three significant structural break dates are identified which coincide with the recovery period of major economic and political events such as the Asian Financial Crisis in 1997, the stock market plunge in China, Japan and Europe in 2007, and the 2008 Global Financial Crisis. In addition, results of the ARCH model indicate consistent negative and statistically significant coefficients for three break dummies across ARCH specifications. The negative coefficients suggest a decline in conditional volatility following each structural break, implying that these events marked transitions into relatively more stable periods rather than heightened market turbulence. This study emphasises the significance of accounting for structural breaks in volatility modelling. The inclusion of breakpoints improves model predictability, offering better insights for investors, policymakers, and analysts concerning the market behaviour under evolving economic conditions in the Malaysian market.