1-s2.0-S0378437119300676-main.pdf (1.6 MB)
Modeling non-stationarities in high-frequency financial time series
journal contribution
posted on 2023-06-09, 16:35 authored by Linda Ponta, Mailan Trinh, Marco Raberto, Enrico Scalas, Silvano CincottiWe study tick-by-tick financial returns for the FTSE MIB index of the Italian Stock Exchange (Borsa Italiana). We confirm previously detected non-stationarities. Scaling properties reported before for other high-frequency financial data are only approximately valid. As a consequence of our empirical analyses, we propose a simple model for non-stationary returns, based on a non-homogeneous normal compound Poisson process. It turns out that our model can approximately reproduce several stylized facts of high-frequency financial time series. Moreover, using Monte Carlo simulations, we analyze order selection for this class of models using three information criteria: Akaike’s information criterion (AIC), the Bayesian information criterion (BIC) and the Hannan-Quinn information criterion (HQ). For comparison, we perform a similar Monte Carlo experiment for the ACD (autoregressive conditional duration) model. Our results show that the information criteria work best for small parameter numbers for the compound Poisson type models, whereas for the ACD model the model selection procedure does not work well in certain cases.
History
Publication status
- Published
File Version
- Accepted version
Journal
Physica A: Statistical Mechanics and its ApplicationsISSN
0378-4371Publisher
ElsevierExternal DOI
Volume
521Page range
173-196Department affiliated with
- Mathematics Publications
Research groups affiliated with
- Probability and Statistics Research Group Publications
Full text available
- Yes
Peer reviewed?
- Yes
Legacy Posted Date
2019-01-21First Open Access (FOA) Date
2020-01-19First Compliant Deposit (FCD) Date
2019-01-21Usage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC