TONG Hengqing, . RATES OF CONVERGENCE IN EMPIRICAL BAYESIAN ESTIMATION OF COEFFICIENT AND ERROR VARIANCE IN LINEAR REGRESSION[J]. Chinese Journal of Applied Probability and Statistics, 1990, 6(3): 242-248.
Citation:
TONG Hengqing, . RATES OF CONVERGENCE IN EMPIRICAL BAYESIAN ESTIMATION OF COEFFICIENT AND ERROR VARIANCE IN LINEAR REGRESSION[J]. Chinese Journal of Applied Probability and Statistics, 1990, 6(3): 242-248.
TONG Hengqing, . RATES OF CONVERGENCE IN EMPIRICAL BAYESIAN ESTIMATION OF COEFFICIENT AND ERROR VARIANCE IN LINEAR REGRESSION[J]. Chinese Journal of Applied Probability and Statistics, 1990, 6(3): 242-248.
Citation:
TONG Hengqing, . RATES OF CONVERGENCE IN EMPIRICAL BAYESIAN ESTIMATION OF COEFFICIENT AND ERROR VARIANCE IN LINEAR REGRESSION[J]. Chinese Journal of Applied Probability and Statistics, 1990, 6(3): 242-248.
In this paper, we consider linear regression Y=Xβ+8,8∼N(0,σ2Ik) where β and σ2 are unknown parameters. With general quadratic loss function, we first construct Bayesian estimators of the parameters of multiparameter exponential families. Using multiple kernel function, we construct the EB estimators of the parameters. We obtain their rates of convergence 2(σr−1)/(2r+p+1) closing to 1.