论文信息 - Minimax regret comparison of hard and soft thresholding for estimating a bounded normal mean

Minimax regret comparison of hard and soft thresholding for estimating a bounded normal mean

We study the problem of estimating the mean of a normal distribution with known variance, when prior knowledge specifies that this mean lies in a bounded interval. The focus is on a minimax regret comparison of soft and hard threshold estimators, which have become very popular in the context of wavelet estimation. Under squared-error loss it turns out that soft thresholding is superior to hard thresholding.

Bernd Droge | B. Droge

[1] B. Droge. On finite-sample properties of adaptive least squares regression estimates , 1993 .

[2] C. L. Mallows. Some comments on C_p , 1973 .

[3] W. Strawderman,et al. On the estimation of a restricted normal mean , 1987 .

[4] William E. Strawderman,et al. Admissibility of Estimators of the Mean Vector of a Multivariate Normal Distribution with Quadratic Loss , 1971 .

[5] J. Berger. Statistical Decision Theory and Bayesian Analysis , 1988 .

[6] D. Donoho,et al. Minimax Risk Over Hyperrectangles, and Implications , 1990 .

[7] P. Bickel. Minimax Estimation of the Mean of a Normal Distribution when the Parameter Space is Restricted , 1981 .

[8] G. Casella,et al. Estimating a Bounded Normal Mean , 1981 .

[9] L. Brown. Admissible Estimators, Recurrent Diffusions, and Insoluble Boundary Value Problems , 1971 .

[10] I. Johnstone,et al. Ideal spatial adaptation by wavelet shrinkage , 1994 .

[11] Bernd Droge,et al. Minimax regret analysis of orthogonal series regression estimation : Selection versus shrinkage , 1998 .

[12] Jianqing Fan,et al. Regularization of Wavelet Approximations , 2001 .