zbMATH — the first resource for mathematics

On best linear unbiased estimation in the restricted general linear model. (English) Zbl 0613.62035
T. Mathew [Math. Operationsforsch. Stat., Ser. Stat. 14, 3-6 (1983; Zbl 0531.62061)] gave a necessary and sufficient condition for the BLUE of \(X\beta\) under a general unrestricted linear model \(M=(y,X\beta,V)\) to be also the BLUE of \(X\beta\) under the corresponding restricted model \(M_ r=(y,X\beta |\) \(R\beta =s,V)\). In this note, Mathew’s result is supplemented by a necessary and sufficient condition for the BLUE of \(X\beta\) under \(M_ r\) to be also the BLUE of \(X\beta\) under M. As a corollary, a necessary and sufficient condition is obtained for the two BLUE’s in question to be equal. This corollary shows that a result of J. K. Baksalary and R. Kala [ibid. 10, 27-35 (1979; Zbl 0416.62049)] can be interpreted to be correct, contrary to the claim of Mathew.

62F10 Point estimation
62J05 Linear regression; mixed models
62J99 Linear inference, regression
Full Text: DOI
[1] Baksalary J.K., Math. Operationsforsch. Statist., ser. statist. 10 pp 27– (1979)
[2] DOI: 10.1080/03081087408817070 · doi:10.1080/03081087408817070
[3] Mathew T., Math. Operationsforsch. Statist. ser. statist. 14 pp 3– (1983)
[4] Rao C.R., Sankhya Ser. A 33 pp 371– (1971)
[5] Rao C.R., Generalized Inverse of Matrices and its Applications (1971) · Zbl 0236.15004
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.