메뉴 건너뛰기
소속 기관 / 학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
고객센터 ENG
주제분류

논문 기본 정보

저자정보
(국립한국해양대학교) (국립한국해양대학교)
저널정보
한국무역통상학회 무역통상학회지 무역통상학회지 제26권 제5호
오류 신고하기

피인용 0

검색

    초록·키워드

    This study investigates conditional heteroskedasticity, asymmetric volatility, and out-of-sample forecasting performance in the Shanghai Containerized Freight Index (SCFI). Using 795 weekly observations from January 2011 to March 2026, we estimate ARIMA-EGARCH and ARIMA-GJR-GARCH models with Student’s t innovations. The results reveal significant conditional heteroskedasticity and asymmetric volatility, with negative shocks exerting stronger effects than positive shocks. However, the asymmetry is sensitive to estimation methods and sample periods, particularly when excluding the COVID-19 pandemic. Both models exhibit near-unit volatility persistence, providing insufficient evidence of mean reversion. In a 13-week out-of-sample evaluation, ARIMA-EGARCH achieves the lowest forecasting error, although differences across models are small. A 104-week rolling evaluation further indicates that both GARCH-based models significantly outperform ARIMA in terms of squared forecast errors, while their relative performance is statistically indistinguishable. Because the models share the same mean specification, these improvements cannot be attributed specifically to asymmetric volatility modeling. Overall, the findings highlight persistent and potentially asymmetric freight-rate volatility while emphasizing the importance of model specification and sample selection.

    최근 본 자료 전체보기