Unbiased invariant minimum norm estimation in generalized growth curve model. (English) Zbl 1112.62054

Summary: This paper considers the generalized growth curve model \(Y= \sum^m_{i=1}X_iB_i Z_i'+U{\mathcal E}\) subject to \[ R(X_m)\subseteq R(X_{m-1})\subseteq\cdots\subseteq R(X_1), \] where \(B_i\) are the matrices of unknown regression coefficients, \(X_i, Z_i\) and \(U\) are known covariate matrices, \(i=1,2,\dots,m\), and \({\mathcal E}\) splits into a number of independently and identically distributed subvectors with mean zero and unknown covariance matrix \(\Sigma\). An unbiased invariant minimum norm quadratic estimator \((MINQE(U,I))\) of \(\text{tr}(C \Sigma)\) is derived and the conditions for its optimally under the minimum variance criterion are investigated. The necessary and sufficient conditions for \(MINQE(U,I)\) of \(\text{tr}(C\Sigma)\) to be a uniformly minimum variance invariant quadratic unbiased estimator \((UMVIQUE)\) are obtained. An unbiased invariant minimum norm quadratic plus linear estimator \((MINQLE(U,I))\) of \(\text{tr}(C\Sigma)+\sum^m_{i=1}\text{tr}(D_i'B_i)\) is also given. To compare with the existing maximum likelihood estimator (MLE) of \(\text{tr}(C\Sigma)\), we conduct some simulation studies which show that our proposed estimator performs very well.


62H12 Estimation in multivariate analysis
62J99 Linear inference, regression
62J05 Linear regression; mixed models
Full Text: DOI


[1] Drygas, H., The estimation of residual variance in regression analysis, Math. oper. statis., 3, 373-388, (1972) · Zbl 0255.62054
[2] Drygas, H., Best quadratic unbiased estimation in variance – covariance component models, Math. oper. statis., 8, 211-231, (1977) · Zbl 0371.62094
[3] Drygas, H., \(\textit{MINQE}\)-theory and the estimation of residual variance in regression analysis, Statistics, 19, 341-347, (1988) · Zbl 0649.62058
[4] Eaton, M.L., Gauss – markov estimation for multivariate linear models: a coordinate-free approach, Ann. math. statis., 41, 528-538, (1970) · Zbl 0195.20101
[5] Kleffe, J., Simultaneous estimation of expectation and covariance matrix in linear models, Math. oper. statis., 9, 443-478, (1978) · Zbl 0415.62026
[6] Kleffe, J., On Hsu’s theorem in multivariate regression, J. multivariate anal., 9, 442-452, (1979) · Zbl 0435.62052
[7] Pan, J.X.; Fang, K.T., Growth curve models and statistical diagnostics, (2002), Springer New York · Zbl 1024.62025
[8] Potthoff, R.F.; Roy, S.N., A generalized multivariate analysis of variance model useful especially for growth curve problems, Biometrika, 51, 313-326, (1964) · Zbl 0138.14306
[9] Rao, C.R.; Kleffe, J., Estimation of variance components and applications, (1988), North-Holland New York · Zbl 0645.62073
[10] Rosen, D.V., Maximum likelihood estimation in multivariate linear normal models, J. multivariate anal., 31, 187-200, (1989) · Zbl 0686.62037
[11] Seely, J.; Zyskind, G., Linear spaces and minimum variance unbiased estimation, Ann. math. statis., 42, 691-703, (1971) · Zbl 0217.51602
[12] Xu, C.Y.; Yang, W.L.; Jiang, W.J., Estimation of parameters in growth curve model, (1996), Beijing Normal University Press Beijing
[13] Yang, W.L., \(\textit{MINQE}(U, I)\) and \(\textit{UMVIQUE}\) of the covariance matrix in the growth curve model, Statistics, 26, 49-59, (1995)
[14] Yang, W.L., The simultaneous estimation of \(\operatorname{tr}(D^\prime B) + \operatorname{tr}(C \operatorname{\Sigma})\) in the growth curve model, Science in China (ser.A), 24, 474-484, (1995)
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.