Institutional Repository of Vilnius University
Multifractality of high frequency data and volatility forecasting /
Abstract
dc:descriptionThe purpose of this thesis is to emphasize the importance of multi-fractal concept by providing an empirical evidence using intra-day financial time series. The Multifractal Detrended Fluctuation analysis and the Generalized Hurst exponent methods were applied on the price indices of Dow Jones Industrial Average, Australian Securities Exchange, Nikkei-225 and NASDAQ-100. The presence of multi-scaling was detected on prices of each mentioned index. In order to employ high frequency data we calculated realized volatility and applied the Binomial Markov-Switching Multifractal model. The power of prediction accuracy was compared to well-established Heterogeneous Auto-Regressive model. The Binomial Markov-Switching Multifractal model showed better performance on the short horizons while Heterogeneous Auto-Regressive model surpassed the latter on long horizon.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2018
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Valaitis, Agnius,
- Contributors dc:contributor
-
- Norvaiša, Rimas
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD35436584&prefLang=en_US