The strong uniform convergence of multivariate variable kernel estimates

From MaRDI portal





The authors investigate a kernel density estimator \(f_ n(x)\) for a multivariate uniformly continuous density f(x) with local bandwidth given by the distance to the kth nearest neighbour where k/log \(n\to \infty\) which had been introduced by \textit{L. Breiman}, \textit{W. Meisel} and \textit{E. Purcell} [Technometrics 19, 135-144 (1977; Zbl 0379.62023)] and \textit{T. J. Wagner} [IEEE Trans. Inf. Theory IT-21, 438-440 (1975; Zbl 0312.62033)]. They show \[ \lim_{n}\sup_{x}| f_ n(x)-f(x)| =0 \] for the kernel being the indicator function of a ball and generalize this to kernels of the type \(K(x)=L(\| x\|)\), where the support of L is [0,1] and L is nonincreasing and bounded.











This page was built for publication: The strong uniform convergence of multivariate variable kernel estimates

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4723025)