Regression models with time series errors (Q2703244)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Regression models with time series errors |
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Regression models with time series errors |
scientific article |
Statements
1 March 2001
0 references
autocorrelation
0 references
polynomial splines
0 references
semiparametric regression
0 references
ARMA models
0 references
additive regression
0 references
autoregressive model fitting
0 references
Regression models with time series errors (English)
0 references
It is shown that many of the standard asymptotic results for autocorrelations, partial autocorrelations and parameters of a stationary time series continue to hold for residuals from nonparametric, semiparametric or additive regression under appropriate conditions on the regression functions, error processes and the linear smoothers. In models of the form NEWLINE\[NEWLINEY_{t}=r(X_{t})+Z_{t},NEWLINE\]NEWLINE where \(r\) is an unknown function and \(X_{t}\) is a covariate process independent of the stationary error \(Z_{t},\) conditions under which estimators based on residuals \(\hat Z _{1},\dots,\hat Z_{n},\) obtained from linear smoothers, are asymptotically equivalent to those based on the actual errors \(Z_{1},\dots, Z_{n},\) are given.
0 references
