Search
Abstract
When identifying a nonstationary autoregressive process, e.g. for the purpose of signal prediction or parametric spectrum estimation, two important decisions must be taken. First, one should choose the appropriate order of the autoregressive model, i.e., the number of autoregressive coefficients that will be estimated. Second, if identification is carried out using the local estimation technique, such as the localized version of the method of least squares, one should select the most appropriate estimation bandwidth, i.e., the effective width of the local data window used for the purpose of parameter tracking. The paper presents the first unified treatment of the problem of joint order and bandwidth selection. Two solutions to this problem are examined, first based on the predictive least squares principle, and second exploiting the suitably modified Akaike’s final prediction error statistic. It is shown that the best results are obtained if the two approaches mentioned above are combined.
Citations
-
0
CrossRef
-
0
Web of Science
-
0
Scopus
Authors (2)
Cite as
Full text
- Publication version
- Accepted or Published Version
- License
- Copyright (EURASIP 2017)
Keywords
Details
- Category:
- Conference activity
- Type:
- materiały konferencyjne indeksowane w Web of Science
- Title of issue:
- 25th European Signal Processing Conference (EUSIPCO 2017) strony 1505 - 1509
- Language:
- English
- Publication year:
- 2017
- Bibliographic description:
- Niedźwiecki M., Ciołek M..: On joint order and bandwidth selection for identification of nonstationary autoregressive processes, W: 25th European Signal Processing Conference (EUSIPCO 2017), 2017, ,.
- DOI:
- Digital Object Identifier (open in new tab) 10.23919/eusipco.2017.8081451
- Verified by:
- Gdańsk University of Technology
seen 129 times