Title: Forecasting non-stationary financial data with OIIR-filters and composed threshold models
Authors : Wildi, Marc
Proceedings: Decision technologies for computational finance : proceedings of the Fifth International Conference Computational Finance
Pages : 391
Pages to: 402
Conference details: 5th International Conference Computational Finance, London, 15-17 December 1997
Publisher / Ed. Institution : Springer
Publisher / Ed. Institution: Boston
Issue Date: 1998
License (according to publishing contract) : Licence according to publishing contract
Series : Advances in computational management science
Series volume: 2
Type of review: Peer review (abstract)
Language : English
Subjects : High pass filter; Stationary component; Filter design; Infinite impulse response; Amplitude function
Subject (DDC) : 332: Financial economics
500: Natural sciences and mathematics
Abstract: The paper proposes a forecasting-technique well suited to stationary and non-stationary economic or financial data. Two methods are used which together generalize the Box-Jenkins ARIMA-technique: Optimized-Infinite-Impulse-Response-Filters generalize difference-filters and composed-threshold (piecewise linear) models generalize linear ARMA-models.
Departement: School of Engineering
Publication type: Conference paper
DOI : 10.1007/978-1-4615-5625-1_31
ISBN: 978-0-7923-8309-3
978-1-4615-5625-1
ISSN: 1388-4301
URI: https://digitalcollection.zhaw.ch/handle/11475/16835
Appears in Collections:Publikationen School of Engineering

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.