뉴스가 중국과 일본 방한객의 변동성에 미치는 충격의 추정 (Measuring the Impact of News on Volatility of Chinese and Japanese Visitors to Korea)

서지정보

· 발행기관 : 한국호텔관광학회
· 수록지 정보 : 호텔관광연구 / 17권 / 3호 / 334 ~ 350페이지
· 저자명 : 모수원, 이광배

초록

The importance of a correctly specified volatility model is clear from the range of applications requiring estimates of conditional volatilities. That is because volatility can be regarded as a measure of information flow. We employ three models of predictable volatility and present a news impact curve which characterizes the impact of past shocks on the visitor volatility implicit in a volatility model. This paper suggests several diagnostic tests based on the news impact curve and compares the GARCH model with other volatility models that allow for asymmetry in the impact of news on volatility. The unpredictable visitors indicate empirical distributions with heavy tails relative to the normal distribution. There is some negative skewness and zero excess kurtosis is confidently rejected. The kernel density function also does not show normal distributions. From the Ljung-Box test statistic for twelfth-order serial correlation, we find significant serial correlation left in the visitor series after our adjustment procedure. The sign bias test and size bias test are highly significant. These statistics mean that GARCH model is appropriate for estimating volatility. We find that both EGARCH and GJR volatility model are useful approaches to modeling conditional heteroskedasticity of the Chinese and Japanese visitors to Korea. There is, however, a difference between Chinese and Japanese responses to news. Chinese visitors tend to respond more sensitive to good news rather than bad news, while Japanese visitors react greater to bad news.

영어초록

The importance of a correctly specified volatility model is clear from the range of applications requiring estimates of conditional volatilities. That is because volatility can be regarded as a measure of information flow. We employ three models of predictable volatility and present a news impact curve which characterizes the impact of past shocks on the visitor volatility implicit in a volatility model. This paper suggests several diagnostic tests based on the news impact curve and compares the GARCH model with other volatility models that allow for asymmetry in the impact of news on volatility. The unpredictable visitors indicate empirical distributions with heavy tails relative to the normal distribution. There is some negative skewness and zero excess kurtosis is confidently rejected. The kernel density function also does not show normal distributions. From the Ljung-Box test statistic for twelfth-order serial correlation, we find significant serial correlation left in the visitor series after our adjustment procedure. The sign bias test and size bias test are highly significant. These statistics mean that GARCH model is appropriate for estimating volatility. We find that both EGARCH and GJR volatility model are useful approaches to modeling conditional heteroskedasticity of the Chinese and Japanese visitors to Korea. There is, however, a difference between Chinese and Japanese responses to news. Chinese visitors tend to respond more sensitive to good news rather than bad news, while Japanese visitors react greater to bad news.

참고자료

· 없음