Analysis on the Forecasting Performance of KOSPI200 Volatility between Long Memory Model and Regime-Switching Model (in Korean)
Joonhyuk Song ()
Additional contact information
Joonhyuk Song: Division of Economics, Hankuk University of Foreign Studies
Economic Analysis (Quarterly), 2011, vol. 17, issue 4, 99-127
Abstract:
This paper investigates the long memory property of KOSPI200 volatility and compare the forecasting power between the long memory and regime switching models. When no regime-switching is assumed, the realized volatility and VKOSPI show no signs of permanent memory, but suggest that both of them have stationary long memory properties. Long memory can be spurious in the presence of structural or regime changes. In order to see this, 1-day and 5-day ahead volatility forecasting test, based on Diebold-Mariano(1995), are conducted to compare model prediction power between long memory and regime-switching models. The test shows that regime-switching model is more accurate in 1-day ahead realized volatility forecasting, while 1-day ahead forecasting shows no difference between the models. It suggests that long memory process of stock volatility can be better understood when structural-changing features of the volatility are given more attention.
Keywords: Realized volatility; VKOSPI; Long memory; DM test (search for similar items in EconPapers)
JEL-codes: G10 G13 G17 (search for similar items in EconPapers)
Date: 2011
References: Add references at CitEc
Citations:
Downloads: (external link)
http://imer.bok.or.kr/attach/imer_kor/2545/2013/12/1386309774519.pdf (application/pdf)
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:bok:journl:v:17:y:2011:i:4:p:99-127
Access Statistics for this article
Economic Analysis (Quarterly) is currently edited by Wook Sohn, Hwan-koo Kang and Jaerang Lee
More articles in Economic Analysis (Quarterly) from Economic Research Institute, Bank of Korea Contact information at EDIRC.
Bibliographic data for series maintained by Economic Research Institute ().