Publication type: Article in scientific journal
Type of review: Peer review (publication)
Title: About model-based time series procedures : some remarks to TRAMO/SEATS and CENSUS X-12-ARIMA
Authors: Stier, W.
Wildi, Marc
Published in: Allgemeines Statistisches Archiv
Volume(Issue): 2002
Issue: 86
Page(s): 447
Pages to: 458
Issue Date: 2002
Publisher / Ed. Institution: Springer
ISSN: 0002-6018
Language: English
Subject (DDC): 500: Natural sciences
Abstract: Model-based approaches are commonly used in extracting signals of interest in time series which are buried in "noise". In order to avoid boundary problems due to the usually preferred symmetric extraction filters (no phase-shift between filter-input and filter-output), finite data are extended by fore- and backcasts generated by a model of the series to be analyzed. We show that despite this the resulting filters may become heavily asymmetric. Moreover, different identification and/or estimation procedures may lead to different models resulting in different characteristics of the derived extraction filters. Therefore, the question of "optimality" of these filters arises. Typical examples based on TRAMO/SEATS and X-12-ARIMA are presented.
Fulltext version: Published version
License (according to publishing contract): Licence according to publishing contract
Departement: School of Engineering
Organisational Unit: Institute of Data Analysis and Process Design (IDP)
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.