본 연구는 한국, 미국, 일본, 영국, 프랑스, 독일의 주별 주식수익률을 선정하고 다변량 EGARCH-M모형을 통해 조건부평균과 조건부분산을 이용하여 수익률과 변동성의 전이효과를 파악하고자 한다. 본 연구를 요약하면, 첫째, 조건부수익률의 결과, 과거의 미국시장의 수익률의 잔차가 현재의 한국과 영국 및 독일시장으로 모두 양(+), 현재의 자체내 미국시장으로 유의한 음(-)의 조건부수익률 전이효과를 나타냈고, 과거의 일본시장에서 현재의 자체내 일본시장으로 양(+)과 현재의 한국과 미국시장으로 유의한 음(-)의 수익률 전이효과를 나타내었다. 또한 과거의 영국시장에서 현재의 미국과 일본시장으로, 과거의 독일시장에서 현재의 한국과 미국시장으로 모두 유의한 양(+)의 수익률 전이효과를 나타냈다. 또한 한국을 제외하고 미국과 프랑스 및 독일시장에 대한 기대수익률과 조건부변동성간 관련성이 유의하여 매우 높았다. 둘째, 조건부변동성의 결과, 과거의 미국시장의 변동성충격이 한국 등 5개 시장을 제외하고 현재의 프랑스시장의 변동성에만 유의한 양(+)의 변동성 전이효과를 나타내었으며, 또한 과거의 한국시장의 변동성의 잔차(충격)가 미국시장만을 제외하고, 현재의 한국, 영국, 프랑스, 독일시장의 변동성에 유의한 음(-)의 조건부변동성 전이효과를 나타내었고, 또한 과거의 일본시장의 변동성충격이 한국과 미국시장 등을 제외하고 현재의 프랑스, 일본과 영국시장의 변동성에 양(+)과 음(-)의 유의한 변동성 전이효과를 보여 주었다. 셋째, 전이효과검정후의 잔차분석의 결과, 왜도는 모두 음(-)의 비대칭이었고 독일 등 5개 시장의 첨도는 정규분포에 가까웠으며 한국, 미국, 영국시장은 안정적이면서 단위근이 없어 모든 시장이 안정적이고 LM검정에서도 한국, 미국, 일본시장은 안정적이었다. 따라서 본 연구에서 주요 선진국들 중 미국과 일본 및 독일시장에서만 현재의 한국시장으로 조건부수익률 전이효과를 나타냈고, 또한 미국과 일본시장을 제외하고 과거의 영국과 프랑스 및 독일시장에서만 현재의 한국시장으로 조건부변동성 전이효과를 나타내었다. 반대로 과거의 한국시장에서 현재의 미국과 일본시장으로 수익률과 변동성 전이효과가 전혀 존재하지 않았다. 또한 악재인 음(-)의 뉴스충격이 호재인 양(+)의 잔차보다 다른 시장의 변동성을 더 증가시킨다는 레버리지효과와 비대칭효과가 미국과 영국을 제외하고 한국과 일본, 프랑스 및 독일시장에서만 존재하였으며, 또한 한국과 영국을 제외하고 미국, 일본, 프랑스, 독일시장에서 변동성의 지속성이 매우 높았다.

The purpose of this study is to examine the spillover effects of return and volatility in the conditional mean and conditional variance equation through a multivariate EGARCH-M model, selecting weekly stock returns in Korea, the U.S., Japan, the U.K., France, and Germany. The results of this study can be outlined as follows: First, in conditional return equation, the analysis showed positive conditional return spillover effects in that past residual return in the stock market of Korea has statistically significant effects on current returns in the stock markets of the U.K. and France. It was found that past returns in the U.S. has positive spillovers on current returns in the stock markets of Korea, the U.K., and Germany, and also has significant negative spillovers on own current returns. And it was found that past returns of Japan has positive spillovers on own current returns, but has significant negative spillovers on current returns of stock markets in Korea and the U.S.. In particular, there was significant relationship between expected returns and conditional volatility in the U.S., France, and Germany. Second, in conditional volatility equation, the analysis showed negative conditional volatility spillover effects, in which past residual shocks of volatility in Korean stock market has statistically significant effects on current volatility in Korea, the U.K., France, and Germany, except the U.S.. It was found that past volatility impacts in the U.S. has significant positive spillovers on current volatility in France only, except other stock markets including Korea. In addition, it was found that past volatility impacts in Japan has significant positive spillovers on current volatility in France and has significant negative spillovers upon current volatility in Japan and the U.K., except Korea and the U.S.. And it was found that past volatility impacts in the U.K. has significant negative spillovers upon current volatility in Korea, Japan, and Germany, while past volatility shocks in France has significant positive spillovers on current volatility in the U.S., the U.K., and France, but has significant negative spillovers on current volatility in Korea. Except the U.S. and the U.K., it was notable that Korean, Japanese, French and German stock market all showed significant negative leverage effects and asymmetric effects. Except Korea and the U.K., it was noted that other 4 major stock markets all showed statistically significant persistence of volatility. Third, in residual analysis, it was found that all markets showed negative skewness, and other major stock markets except Germany showed kurtosis equivalent to normal distribution, but didn't show any normal distribution in Jarque-Bera statistic. For modified Q and Q2 statistic, stock markets of Korea, the U.S., and the U.K. had serially uncorrelated and stable, showing there was no unit root in all markets. In LM test, Korea, the U.S., Japan, and the U.K. were all in stable conditions. Hence, this study came to a conclusion that past impacts of return and volatility in Korean stock market had no spillover effects upon current return and volatility in the U.S. and Japan, while past returns in the U.S., Japan, and Germany had spillover effects on current returns in Korean, and past volatility impacts in the U.K., France, and Germany had current volatility spillover effects in Korea. And, Korean, Japanese, French, and German stock markets showed leverage and asymmetric effects.

The purpose of this study is to examine the spillover effects of return and volatility in the conditional mean and conditional variance equation through a multivariate EGARCH-M model, selecting weekly stock returns in Korea, the U.S., Japan, the U.K., France, and Germany. The results of this study can be outlined as follows: First, in conditional return equation, the analysis showed positive conditional return spillover effects in that past residual return in the stock market of Korea has statistically significant effects on current returns in the stock markets of the U.K. and France. It was found that past returns in the U.S. has positive spillovers on current returns in the stock markets of Korea, the U.K., and Germany, and also has significant negative spillovers on own current returns. And it was found that past returns of Japan has positive spillovers on own current returns, but has significant negative spillovers on current returns of stock markets in Korea and the U.S.. In particular, there was significant relationship between expected returns and conditional volatility in the U.S., France, and Germany. Second, in conditional volatility equation, the analysis showed negative conditional volatility spillover effects, in which past residual shocks of volatility in Korean stock market has statistically significant effects on current volatility in Korea, the U.K., France, and Germany, except the U.S.. It was found that past volatility impacts in the U.S. has significant positive spillovers on current volatility in France only, except other stock markets including Korea. In addition, it was found that past volatility impacts in Japan has significant positive spillovers on current volatility in France and has significant negative spillovers upon current volatility in Japan and the U.K., except Korea and the U.S.. And it was found that past volatility impacts in the U.K. has significant negative spillovers upon current volatility in Korea, Japan, and Germany, while past volatility shocks in France has significant positive spillovers on current volatility in the U.S., the U.K., and France, but has significant negative spillovers on current volatility in Korea. Except the U.S. and the U.K., it was notable that Korean, Japanese, French and German stock market all showed significant negative leverage effects and asymmetric effects. Except Korea and the U.K., it was noted that other 4 major stock markets all showed statistically significant persistence of volatility. Third, in residual analysis, it was found that all markets showed negative skewness, and other major stock markets except Germany showed kurtosis equivalent to normal distribution, but didn't show any normal distribution in Jarque-Bera statistic. For modified Q and Q2 statistic, stock markets of Korea, the U.S., and the U.K. had serially uncorrelated and stable, showing there was no unit root in all markets. In LM test, Korea, the U.S., Japan, and the U.K. were all in stable conditions. Hence, this study came to a conclusion that past impacts of return and volatility in Korean stock market had no spillover effects upon current return and volatility in the U.S. and Japan, while past returns in the U.S., Japan, and Germany had spillover effects on current returns in Korean, and past volatility impacts in the U.K., France, and Germany had current volatility spillover effects in Korea. And, Korean, Japanese, French, and German stock markets showed leverage and asymmetric effects.

volatility, multivariate, spillover, leverage, asymmetric effects