Premium
Reparametrization of autoregressive models with coefficients depending on covariables: application to EEG spectrum maturation
Author(s) -
Clark I.,
Biscay R.,
Jiménez J. C.
Publication year - 1997
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/(sici)1097-0258(19970815)16:15<1745::aid-sim598>3.0.co;2-6
Subject(s) - autoregressive model , regression , statistics , econometrics , range (aeronautics) , electroencephalography , normative , regression analysis , series (stratigraphy) , computer science , transformation (genetics) , mathematics , psychology , paleontology , philosophy , materials science , epistemology , psychiatry , composite material , biology , biochemistry , chemistry , gene
To describe the spectral characteristics of the EEG development through autoregressive (AR) time series models it is necessary to perform regression analysis of the AR parameters with regards to the age of the subject. A major difficulty in this approach is the very complex nature of the admissible region of the AR coefficients, which impedes the straight use of regression techniques. The present paper overcomes this difficulty by first applying the Barndorff–Nielsen and Schou reparametrization of AR models, followed by Fisher's transformation, and then carrying out age regression analysis of the transformed parameters. We apply this approach to real EEG data obtained from a normative sample of subjects in the age range from 5 to 95 years. © 1997 by John Wiley & Sons, Ltd.