Baillie의 ARFIMA, FIVECM 및 Full BEKK-GARCH를 이용한 중국과 홍콩주식시장간의 수익률과 변동성의 전이효과에 대한 연구

This study is to analyze both bivariate conditional mean and conditional volatility spillover effects in order to examine relationship - i.e. market efficiency, long-run equilibrium, short-run adjustments and spillover effects - between Hong Kong(HK) and Chinese A(SHA & SZA), B(SHB & SZB) stock market using a combination of ARFIMA, FIVECM and Full BEKK-GARCH model. The results of this study can be outlined as follows: First, both ARFIMA(as a long memory model on returns) and FIGARCH model(as a long memory model on volatility) had a long memory in individual markets and cointegration residuals in 6 markets. Thus stock markets was inefficient. Second, it was notable that SHB market was more rapidly adjusted to recover from equilibrium with HK than SHA. Likewise, SZB market was more rapidly adjusted to recover from equilibrium with HK than SZA. Price discovery process of SHB and SZB related to HK was more efficient than that of SHB and SZB market related to A-share. Third, there were conditional homoskedasticity, so stock markets were almost efficient. Therefore, this study was confirmed that there were unidirectionally conditional volatility spillover effects from two B-share to HK, from SZB to SZA, from HK to SHA as well as directionally between SHB and SHA.