Simultaneous empirical Bayes estimation for mean and error variance in normal distribution

doi:10.7523/j.issn.2095-6134.2013.04.005

Abstract

Abstract:

In the normal distribution, by assuming that the parameters of mean and error variance have the normal inverse-Gamma distribution, the simultaneous Bayes estimators of mean and error variance are derived. By using historical samples, the parametric empirical Bayes estimators (PEBE) are constructed. Superiorities of PEBE over UMVUE in mean parameter and error variance are obtained in terms of the MSE criterion. Finally, we give a remark on the main results.

Key words: normal distribution, mean parameters, error variance, parametric empirical Bayes estimation (PEBE), mean-square error (MSE) criterion

CLC Number:

O212.1

QIU Li-Sha, WEI Lai-Sheng. Simultaneous empirical Bayes estimation for mean and error variance in normal distribution[J]. , 2013, 30(4): 454-461.

References

[1] Robbins H. An empirical Bayes approach to statistics [C]//Proc Third Berkeley Symp Math Statist Prob. Berkeley: University of California Press, 1956, 1: 157-163.

[2] Robbins H. The empirical Bayes approach to statistical decision problem [J]. Ann Math Statist, 1964, 35: 1-20.

[3] Morris C N. Parametric empirical Bayes inference-theory and application [J]. Journal of the American Statistical Association, 1983, 78: 47-55.

[4] Ghosh M, Saleh A K Md E, Sen P K. Empirical Bayes subset estimation in regression model [J]. Statist Decisions, 1989, 7: 15-35.

[5] Wei L S, Trenkler G. Mean square error matrix superiority of empirical Bayes estimators under misspecification [J]. Test, 1995, 4: 187-205.

[6] Wei L S, Zhang W P. The superiorities of Bayes linear minimun risk estimation in linear model [J]. Comumn Statist Theor Meth, 2007, 36: 917-926.

[7] Singh R S. Empirical Bayes estimation in Lebesgue-expenential families with Rates near the best possible rate [J]. Ann of Statistics, 1979, 7(4): 890-902.

[8] Chen X R. Asymptotically optimal empirical Bayes estimation for parameter of one-dimension discrete exponential families[J]. Chin Ann of Math, 1983, 4B(1): 41-50.

[9] Wang L C, Wei L S. The convergence rates of empirical Bayes estimation of the parameter for scale-exponential family[J]. Chin Ann of Math, 2002, 23A(5): 555-564 (in Chinese). 王立春,韦来生. 刻度指数族参数的经验Bayes估计的收敛速度[J]. 数学年刊, 2002, 23A(5): 555-564.

[10] Berger J O. Statistical decision theory and Bayes analysis [M]. New York: Springer, 1985.

[11] Box G E P, Tiao G C. Bayesian inference in statistical analysis [M]. Massachusetts: Addison-Wesley, 1973.

[12] Tao B. Asymptotically optimal empirical Bayes estimators for the parameters of normal distribution family [J]. J Math Res & Exposition, 1986, 1: 157-162.

[13] Zhang P. Convergence rate of the asymptotic optimal empirical Bayes estimator of the parameters in normal distribution[J]. Systems Science and Mathematical Sciences, 1985, 5(3): 185-191 (in Chinese). 张平. 正态分布参数的渐近最优经验 Bayes估计的收敛速度[J]. 系统科学与数学, 1985, 5(3): 185-191.

[14] Chen L, Wei L S. The superiority of empirical Bayes estimation of the error variance in linear model[J]. Front Math China, 2012, 7(4): 629-644.