Universal pointwise selection rule in multivariate function estimation
Goldenshluger, Alexander ; Lepski, Oleg
Bernoulli, Tome 14 (2008) no. 1, p. 1150-1190 / Harvested from Project Euclid
In this paper, we study the problem of pointwise estimation of a multivariate function. We develop a general pointwise estimation procedure that is based on selection of estimators from a large parameterized collection. An upper bound on the pointwise risk is established and it is shown that the proposed selection procedure specialized for different collections of estimators leads to minimax and adaptive minimax estimators in various settings.
Publié le : 2008-11-15
Classification:  adaptive estimation,  minimax risk,  optimal rates of convergence,  pointwise estimation
@article{1225980575,
     author = {Goldenshluger, Alexander and Lepski, Oleg},
     title = {Universal pointwise selection rule in multivariate function estimation},
     journal = {Bernoulli},
     volume = {14},
     number = {1},
     year = {2008},
     pages = { 1150-1190},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1225980575}
}
Goldenshluger, Alexander; Lepski, Oleg. Universal pointwise selection rule in multivariate function estimation. Bernoulli, Tome 14 (2008) no. 1, pp.  1150-1190. http://gdmltest.u-ga.fr/item/1225980575/