REVISITING THE SUDDEN CHANGES AND VOLATILITY PERSISTENCE IN EUROPEAN CAPITAL MARKETS: SOME EMPIRICAL EVIDENCE

Abstract

Understanding the behaviour of market volatility is crucial for the asset pricing, portfolio selection, risk management, and trading strategies. The standard Generalized Autoregressive Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model assume that there is no shift in variance, hence its inability to produce good estimate of volatility persistence. This research paper re-examines volatility persistence as well as the sudden changes in variance for some major European capital markets. The shifts in variance are detected by iterated cumulative sums of squares (ICSS) algorithm and then incorporated into multivariate BEKK-GARCH model. Information obtained shows that the detected changes correspond to both global and domestic events. Results also showed that volatility persistence is reduced in a controlled volatility change model compares to model ignoring volatility changes. The implication of these results indicates that many previous studies on volatility persistence may have reported overestimated results.