국내 금융시계열의 누적(INTEGRATED)이분산성에 대한 사례분석

Abstract

Conditionally heteroscedastic time series models such as GARCH processes havefrequently provided useful approximations to the real aspects of nancial time series. Itis not uncommon that nancial time series exhibits near non-stationary, say, integratedphenomenon. For stationary GARCH processes, a shock to the current conditionalvariance will be exponentially converging to zero and thus asymptotically negligiblefor the future conditional variance. However, for the case of integrated process, theeect will remain for a long time, i.e., we have a persistent eect of a current shockon the future observations. We are here concerned with providing empirical evidencesof persistent GARCH(1,1) for various fteen domestic nancial time series includingKOSPI, KOSDAQ and won-dollar exchange rate. To this end, kurtosis and Integrated-GARCH(1,1) ts are reported for each data.