STRUCTURAL BREAKS AND GARCH MODELS OF STOCK RETURN VOLATILITY: THE CASE OF VIETNAMESE

Abstract

This study uses the daily rate of return of stock indexes to test the effect of fracture structure. We use ARCH models and GARCH models. Use Zivot-Andrews (for a break point) test and Multiple Breakpoint testing to determine the fracture structure in the GARCH model. The results show that profit margins (ACB and VNINDEX) have deviated and leptokurtic distributions and have no standard distribution. The break in the jar has shown very tight results in time. Backtesting is done by measuring the number of times the loss is greater than the VaR prediction. Verification of the impact of structural fracture suggests that combining structural fractures in the GJR-GARCH model can be used to predict VaR.