Some remarks on general linear model with new regressors
Assume that an original general linear model is misspecified by adding some new regressors. We investigate in such a case relationships between the best linear unbiased estimators under the two models. In particular, we give necessary and sufficient conditions for the best linear unbiased estimators to be equal under the two models.
Year of publication: |
2015
|
---|---|
Authors: | Lu, Changli ; Gan, Shengjun ; Tian, Yongge |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 97.2015, C, p. 16-24
|
Publisher: |
Elsevier |
Subject: | Linear model | Misspecified model | Parametric functions | New regressors | BLUE |
Saved in: