zbMATH — the first resource for mathematics

Different kinds of sufficiency in the general Gauss-Markov model. (English) Zbl 1150.62032
In the general Gauss-Markov model three kinds of sufficiency, i.e., linear sufficiency (LS), invariant linear sufficiency (ILS) and quadratic sufficiency (QS), have been used. It is proved that a transformation is ILS iff it is simultaneously LS and QS.

62J05 Linear regression; mixed models
62B05 Sufficient statistics and fields
Full Text: DOI EuDML
[1] BAKSALARY J. K.-DRYGAS H.: A note on the concepts of sufficiency in the general Gauss-Markov model coordinate-free approach. Forshungsbericht 92/2, 1992.
[2] BAKSALARY J. K.-KALA R.: Linear transformations preserving best linear unbiased estimators in a general Gauss-Markov model. Ann. Statist. 9 (1981) 913-916. · Zbl 0471.62067
[3] DRYGAS H.: Sufficiency and completeness in the general Gauss-Markov model. Sankhua Ser. A 45 (1983), 88-98. · Zbl 0535.62007
[4] MUELLER J.: Sufficiency and completeness in the linear model. J. Multivariate Anal. 21 (1987), 312-323. · Zbl 0614.62080
[5] OKTABA W.-KORNACKI A.-WAWRZOSEK J.: Invariant linearly sufficient transformations of the general Gauss-Markov model. Estimation and testing. Scand. J. Statist. 15 (1988), 117-124. · Zbl 0682.62041
[6] RAO C. R.: Unified theory of linear estimation. Sankhua Ser. A 33 (1971), 371-394. · Zbl 0236.62048
[7] RAO C. R.: Projectors, generalized inverse and the BLUE’s. J. Roy Statist. Soc. Ser. B 36 (1974), 442-448. · Zbl 0291.62077
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.