多元线性混合模型方差分量矩阵的非负估计

马铁丰, 叶仁道, 贾丽杰

数学学报 ›› 2010, Vol. 53 ›› Issue (2) : 349-362.

PDF(640 KB)
PDF(640 KB)
数学学报 ›› 2010, Vol. 53 ›› Issue (2) : 349-362. DOI: 10.12386/A2010sxxb0042
论文

多元线性混合模型方差分量矩阵的非负估计

    马铁丰1, 叶仁道2, 贾丽杰3
作者信息 +

Nonnegative Estimation of Variance Component Matrix in Multivariate Linear Mixed Model

    Tie Feng MA1, Ren Dao YE2, Li Jie JIA3
Author information +
文章历史 +

摘要

本文研究了带有两个方差分量矩阵的多元线性混合模型方差分量矩阵的估计问题.对于平衡模型, 给出了基于谱分解估计的一个方差分量矩阵的非负估计类. 对于非平衡模型, 给出了方差分量矩阵的广义谱分解估计类, 讨论了与ANOVA估计等价的充要条件. 同时, 在广义谱分解估计的基础上给出了一种非负估计类, 并讨论了其优良性. 当具有较小二次风险的非负估计不存在时, 从估计为非负的概率的角度考虑, 将Kelly 和 Mathew (1993)提出的构造具有更小取负值概率的估计类的方法推广到本文的多元模型下, 给出了较谱分解估计相比有更小取负值概率和更小风险的估计类. 最后, 模拟研究和实例分析表明文中理论结果有很好的表现.

 

Abstract

Multivariate linear mixed model with two variance components matrices is considered. For balanced model, based on spectral decomposition estimate (SDE), a class of nonnegative estimate of variance components matrices is proposed. For unbalanced model, a so-called generalized spectral decomposition estimate (GSDE) of variance components matrices is obtained. Our results show a sufficient and necessary condition for the GSDE to be equal to the analysis of variance estimate (ANOVAE). And then, based on GSDE, we also propose a class of nonnegative estimate and discuss the optimality. Furthermore, we extend the method proposed by Kelly and Mathew (1993) to multivariate model in this paper. Finally, some simulations and applications to illustrate the results are presented.

 

关键词

中部塌陷 / 多元线性混合模型 / 方差分量矩阵 / 谱分解估计

Key words

multivariate mixed linear models / variance components matrix / spectral decomposition estimator / central region falling

引用本文

导出引用
马铁丰, 叶仁道, 贾丽杰. 多元线性混合模型方差分量矩阵的非负估计. 数学学报, 2010, 53(2): 349-362 https://doi.org/10.12386/A2010sxxb0042
Tie Feng MA, Ren Dao YE, Li Jie JIA. Nonnegative Estimation of Variance Component Matrix in Multivariate Linear Mixed Model. Acta Mathematica Sinica, Chinese Series, 2010, 53(2): 349-362 https://doi.org/10.12386/A2010sxxb0042

参考文献


[1] Ma T. F., Wang S. G., Estimation of variance components in multivariate mixed linear models with two variance components, Chinese Annals of Mathematics, Ser. A., 2009, 30(1): 73--84 (in Chinese).

[2] Wang S. G., Yin S. J., A new estimate of the parameters in linear mixed models, Science in China, Ser. A., 2002, 10: 434--443 (in Chinese).

[3] Amemiya Y., What should be done when an estimated between-group covariance matrix is not nonnegative definite? Amer. Statist., 1985, 39: 112--117.

[4] Calvin J. A., Dykstra R. L., Maximum likelihood estimation of a set of covariance matrices under Lowner order restriction with applications to balanced multivariate variance components models, Ann. Statist., 1991a, 19: 850--869.

[5] Calvin J. A., REML estimation in unbalanced multivariate variance components models using an EM algorithm, Biometrics, 1993, 49: 691--701.

[6] Calvin J. A., Dykstra R. L., Least squares estimation of covariance matrices in balanced multivariate variance components model. J. Amer. Statist. Assoc., 1991b, 86: 388--395.

[7] Mathew T., Sinha B. K., Sutradhar B., Nonnegative Estimation of Variance Components in General Balanced Mixed Models, In: Saleh (Ed.), Nonparametric Statistics and Related Topics, Amsterdam: Elsevier Science Publishers, 1992, 281--295.

[8] Mathew T., Sinha B. K., Nonnegative Estimation of Individual Variance Components in some Linear Models, a Survey. Statist. Probab (A Raghu Raj Bahadur Festschrift), 1993: 411--444.

[9] Sun Y. J., Sinha B. K., Rosen D. V., Meng Q. Y., Nonnegative estimation of variance components in multivariate unbalanced mixed linear models with two variance components, Journal of Statistical Planning and Inference., 2003, 115: 215--234.

[10] Wu M. X., Wang S. G., A new method of spectral decomposition of covariance mstrix in mixed effects models and its applications, Science in China, Ser. A, 2005, 35: 947--960 (in Chinese).

[11] Shi J. H., Wang S. G., Some properties of spectral decomposition estimate of variance components, Chinese Journal of Applied Probability and Statistics, 2004, 20: 287--294 (in Chinese).

[12] Shi J. H., Wang S. G., The spectral decomposition of covariance matrices for the variance components models, Journal of Multivariate Analysis, 2006, 97: 2190--2205.

[13] Shi J. H., Wang S. G., Generalized spectral decomposition estimate of variance components, Appl Math. J. Chinese Univ. Ser. A, 2005, 20: 83--89 (in Chinese).

[14] Leung P. L., Ng F. Y., Improved estimation of a covariance matrix in an elliptically contoured matrix distribution, Journal of Multivariate Analysis., 2004, 88: 131--137.

[15] Wang S. G., Yang Z. H., Generalized Inverse Matrix and its Application, Beijing: Beijing University of Technology Press, 1996 (in Chinese).

[16] Mathew T., Niyogi A., Sinha B. K., Improved nonnegative estimation of variance components in balanced multivariate mixed models, J. Multivariate Anal, 1994, 51: 83--101.

[17] Kelly R. J., Mathew T., Improved Nonegative Estimation of Variance Components in Some Mixed Model with Unbalanced Data, Journal of the Royal Statistical Society, B, 1993, 55: 897--911.

[18] Yang S. G., Zhu H., Central sinking, financial weakening and studies on realizing the central rising by financial support, Economic Research Journal, 2007, 5: 55--67 (in Chinese).

基金

国家自然科学青年基金(10801005)及国家自然科学基金数学天元青年基金(10926059);西南财经大学211工程三期统计学重点学科建设资助项目

PDF(640 KB)

Accesses

Citation

Detail

段落导航
相关文章

/