@article{oai:ynu.repo.nii.ac.jp:00006834,
 author = {Fotopoulos, S. B.},
 issue = {2},
 journal = {Yokohama Mathematical Journal = 横濱市立大學紀要. D部門, 数学},
 month = {},
 note = {application/pdf, The density estimator ＄f_{￥tau_{n}}(t)=T_{￥overline{n}^{1}}￥sum_{j=1}^{￥tau_{n}}h_{j}^{-1}K((t-X_{j})/h_{j})＄ is considered, where ＄T_{n}＄ is a random index. A sharp rate of convergence for the uniform distance between the probability of ＄f_{￥tau_{n}}(t)＄ , when Properly normed, and the standard normal distribution is obtained. It is shown that for any ＄￥epsilon>0＄ , the cptimum order of the uniform estimate is ＄0(n^{-1/3+￥epsilon})＄ . Similar results have been shown by other authors, but under different assumptions on ＄T_{n}＄ .},
 pages = {89--105},
 title = {ON THE RATE OF CONVERGENCE FOR SEQUENTIAL DENSITY ESTIMATION},
 volume = {39},
 year = {1992}
}