论文信息 - Sharp minimaxity and spherical deconvolution for super-smooth error distributions

Sharp minimaxity and spherical deconvolution for super-smooth error distributions

The spherical deconvolution problem was first proposed by Rooij and Ruymgaart (in: G. Roussas (Ed.), Nonparametric Functional Estimation and Related Topics, Kluwer Academic Publishers, Dordrecht, 1991, pp. 679-690) and subsequently solved in Healy et al. (J. Multivariate Anal. 67 (1998) 1). Kim and Koo (J. Multivauriate Anal. 80 (2002) 21) established minimaxity in the L2-rate of convergence. In this paper, we improve upon the latter and establish sharp minimaxity under a super-smooth condition on the error distribution.

[1] A. Goldenshluger. Density Deconvolution in the Circular Structural Model , 2002 .

[2] L. L. Cam,et al. Asymptotic methods in statistical theory , 1986 .

[3] Jianqing Fan. On the Optimal Rates of Convergence for Nonparametric Deconvolution Problems , 1991 .

[4] Dennis M. Healy,et al. Spherical Deconvolution , 1998 .

[5] F. Ruymgaart,et al. Regularized Deconvolution on the Circle and the Sphere , 1991 .

[6] L. L. Cam,et al. Asymptotic Methods In Statistical Decision Theory , 1986 .

[8] A. Tsybakov,et al. Minimax theory of image reconstruction , 1993 .

[9] Peter T. Kim,et al. Optimal Spherical Deconvolution , 2002 .

[10] A. P. Korostelev,et al. MiniMax Methods for Image Reconstruction , 1993 .

[11] Sam Efromovich,et al. Density Estimation for the Case of Supersmooth Measurement Error , 1997 .