Necessary and sufficient conditions for superiority of misspecified restricted least squares regression estimator
✍ Scribed by Jürgen Groß; Götz Trenkler; Erkki P. Liski
- Publisher
- Elsevier Science
- Year
- 1998
- Tongue
- English
- Weight
- 91 KB
- Volume
- 71
- Category
- Article
- ISSN
- 0378-3758
No coin nor oath required. For personal study only.
✦ Synopsis
Razzaghi [Comm. Statist. Ser. B 16 (1987)
899-902] investigated the di erence of dispersion matrices of competing estimators in the general linear model when the dispersion matrix of the disturbance vector is misspeciÿed. The aim of this paper is to strengthen Razzaghi's result by claiming his su cient condition for nonnegative deÿniteness of the dispersion matrix di erence to be also necessary. A further dispersion matrix di erence is investigated by applying a result of Kabe and Gupta [
📜 SIMILAR VOLUMES
Many processes in automatic regulation, physics, mechanics, biology, economy, ecology, etc. can be modelled by hereditary systems (see, e.g., [1][2][3][4]). One of the main problems for the theory of such systems and their applications is connected with stability (see, e.g., [2][3][4]). Many stabili