ON MISSPECIFICATION OF THE COVARIANCE MATRIX IN LINEAR MODELS

Jian-Ying Rong (P. R. China) and Xu-Qing Liu (P. R. China)

Received December 23, 2007

References:

[1] J. K. Baksalary, S. Puntanen and G. P. H. Styan, A property of the dispersion matrix of the best linear unbiased estimator in the general Gauss-Markov model, Sankhya 52 (1990), 279-296. [2] S. Gnot and M. Grzadziel, Nonnegative minimum biased quadratic estimation in mixed linear models, J. Multivariate Anal. 80 (2002), 217-233. [3] S. Gnot, G. Trenkler and R. Zmyślony, Nonnegative minimum biased quadratic estimation in the linear regression models, J. Multivariate Anal. 54 (1995), 113-125. [4] J. Groß and A. Markiewicz, Characterizations of admissible linear estimators in the linear model, Linear Algebra Appl. 388 (2004), 239-248. [5] C. R. Rao and H. Toutenburg, Linear Models: Least Squares and Alternatives, Springer-Verlag, New York, 1995. [6] Y. Tian and D. P. Wiens, On equality and proportionality of ordinary least squares, weighted least squares and best linear unbiased estimators in the general linear model, Statist. Probab. Lett. 76(12) (2006), 1265-1272. [7] S. G. Wang, Theory of Linear Models and its Applications, Anhui Education Press, China, 1987 (in Chinese). [8] S. Wang and Z. Jia, Inequalities in Matrix Theory, Anhui Education Press, China, 1994 (in Chinese). [9] B. Zhang and X. Zhu, Gauss-Markov and weighted least-squares estimation under a general growth curve model, Linear Algebra Appl. 321 (2000), 387-398.

Keywords and phrases: general linear model, robustness, misspecified model, misspecification, best quadratic unbiased estimation, compound symmetric.