Asymmetry and long-memory volatility: Some empirical evidence using GARCH

摘要

This paper investigates the asymmetry and long-memory volatility behavior of the Malaysian Stock Exchange daily data over a period of 1991-2005. The long-spanning data set enable us to examine piecewise before, during and after the economic crisis encountered in the Malaysian stock market. The daily index returns are adjusted for infrequent trading effect and the estimated Hurst's parameter allows us to rank the market efficiency across the periods. The leverage effect, clustering volatility and long-memory behavior of the volatility are fitted by the asymmetry GARCH models and GARCH with the inclusion of realized volatility at the final period. Across the periods, the results show the mixture of symmetry and asymmetry GARCH modeling. (c) 2006 Elsevier B.V. All rights reserved.