Asymptotic properties in space and time of an estimator in nonlinear functional errors-in-variables models (Q4516147)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1535877
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Asymptotic properties in space and time of an estimator in nonlinear functional errors-in-variables models |
scientific article; zbMATH DE number 1535877 |
Statements
Asymptotic properties in space and time of an estimator in nonlinear functional errors-in-variables models (English)
0 references
27 November 2000
0 references
errors in covariates
0 references
weak dependence
0 references
consistency
0 references
asymptotic normality
0 references
0.8812338709831238
0 references
0.8535634279251099
0 references
0.8327544927597046
0 references
A nonlinear regression model with errors in covariates is considered of the form \[ y_i= g(\xi_i,\beta_0)+\delta_i,\qquad x_i=\xi_i+\varepsilon_i, \] where \((y_i,x_i)\), \(i=1,\dots,n\), are the observations, \(\xi_i\) are nonrandom unobserved nuisance parameters, \(\beta_0\) is a parameter to be estimated, \(g\) is some known function, and \(\varepsilon_i\), \(\delta_i\) are random errors with zero mean. The sequence \(\{\varepsilon_i\}\) is independent on \(\{\delta_i\}\), and the sequences of errors satisfy some strong mixing conditions. The estimator \(\hat\beta_n\) for \(\beta_0\) is constructed as a minimizer of the criterion \[ Q_n(\beta)=n^{-1}\sum_{i=1}^n(\|y_i-f_i(x_i,\beta)\|^2+ h_i(x_i,\beta)-\|f_i(x_i,\beta\|^2), \] where the functions \(f_i\) and \(g_i\) are such that \[ Ef_i(\xi+\varepsilon_i,\beta)=g(\xi,\beta),\quad E h_i(\xi+\varepsilon_i,\beta)=\|g(\xi,\beta)\|^2, \] for all possible \(\xi\) and \(\beta\). This criterion can be considered as a least squares criterion improved for the errors in covariates model. Consistency and asymptotic normality of \(\hat\beta_n\) are proved under some regularity conditions.
0 references