|
ИСТИНА |
Войти в систему Регистрация |
Интеллектуальная Система Тематического Исследования НАукометрических данных |
||
Long Memory and Periodicity in Intraday Volatilityстатья
- Авторы: Rossi Eduardo, Fantazzini Dean
- Журнал: Journal of Financial Econometrics
- Год издания: 2014
- Издательство: Oxford University Press
- Местоположение издательства: United Kingdom
- DOI: 10.1093/jjfinec/nbu006
- Аннотация: Intraday return volatility is characterized by the contemporaneous presence of periodicity and long memory. This article proposes two new parameterizations of the intraday volatility process that account for both features: the Fractionally Integrated Periodic EGARCH and the Seasonal Fractional Integrated Periodic EGARCH. The analysis of hourly E-mini S&P 500 futures returns shows that the volatility is characterized by a statistically significant long-range dependence coupled with a periodic leverage effect, with negative return shocks having a larger effect on volatility during the US trading period. Long memory estimates obtained with nonperiodic long memory models are greater than those obtained with FI-PEGARCH and SFI-PEGARCH models. A simulation experiment shows that the long memory component can be strongly biased when periodic patterns are not properly modelled at the intraday level. An out-of-sample forecasting comparison with alternative models shows that a constrained version of the FI-PEGARCH provides superior forecasts.
- Добавил в систему: Фантаццини Деан c.