EGARCH 모형과 GJR-GARCH 모형을 이용한 비대칭 변동성 분석

Abstract

This paper investigate the time series behavior of stock returns for KOSDAQ, KOSPI daily index. The Volatility effects in financial time series vary according to the signs of the return rates. Asymmetric EGARCH and GJR-GARCH processes extend standard GARCH toward asymmetric volatility modeling. As preliminary to detection of asymmetry in volatility, we suggest graphs of squared-log-returns for various financial time series including KOSDAQ, KOSDAQ100, KOSPI, KOSPI200. Next, asymmetric EGARCH(1,1) and GJR-GARCH(1,1) model fits are provided in comparisons with standard GARCH(1,1) models. Analysing these models suggests that asymmetric effect on the conditional volatility of stock return rates certainly exists.