"Asymptotic Properties of the LSE of a Spatial Regression in both Weakly and Strongly Dependent Stationary Random Fields"
We consider asymptotic properties of the least squares estimator(LSE) in spatial regression with correlated errors. Firstly we derive sufficient conditions for the LSE to be strongly consistent and next necessary and/or sufficient conditions tor the LSE to be asymptotically efficient relative to the best liner unbiased estimator(BLUE). Finally we derive the asymptotic distribution of the LSE under conditions on the higher order cumulants of the error terms and the Fourier transforms of the regressors. The main feature is that we propose a unified way in which we can deal with both weakly dependent and strongly dependent error terms.
Year of publication: |
2008-09
|
---|---|
Authors: | Yajima, Yoshihiro ; Matsuda, Yasumasa |
Institutions: | Center for International Research on the Japanese Economy (CIRJE), Faculty of Economics |
Saved in:
Saved in favorites
Similar items by person
-
"On Nonparametric and Semiparametric Testing for Multivariate Time Series"
Yajima, Yoshihiro, (2003)
-
"Covariance Tapering for Prediction of Large Spatial Data Sets in Transformed Random Fields"
Hirano, Toshihiro, (2011)
-
"An Asymptotic Expansion with Push-Down of Malliavin Weights"
Takahashi, Akihiko, (2011)
- More ...