L
Lutz Duembgen
Researcher at University of Bern
Publications - 32
Citations - 657
Lutz Duembgen is an academic researcher from University of Bern. The author has contributed to research in topics: Estimator & Distribution function. The author has an hindex of 10, co-authored 28 publications receiving 585 citations.
Papers
More filters
Posted Content
Optimal Confidence Bands for Shape-Restricted Curves
TL;DR: In this paper, the authors construct confidence bands for convex functions with guaranteed given coverage probability, assuming that the convex function is isotonic or convex, and show that these confidence bands are computationally feasible and shown to be asymptotically sharp optimal.
Posted Content
The density ratio of generalized binomial versus Poisson distributions
Lutz Duembgen,Jon A. Wellner +1 more
TL;DR: In this article, the maximal ratio of the probability that a sum of independent Bernoulli random variables with parameters (p_1, p_2 + p_3 + \cdots) equals a given number of variables is shown to be a function of the weight function of a Poisson distribution.
Posted Content
Local Estimation of a Multivariate Density and its Derivatives
TL;DR: In this paper, four different approaches to estimate a multivariate probability density (or the log-density) and its first and second order derivatives are compared, including local log-likelihood and local Hyvarinen score estimation.
Journal ArticleDOI
Nemirovski's Inequalities Revisited
TL;DR: In this article, the authors compare three different approaches to obtain moment inequalities for arbitrary independent and centered random vectors in Banach spaces: deterministic, type and cotype inequalities, and Bernstein's inequality.
Posted Content
Multiscale Methods for Shape Constraints in Deconvolution: Confidence Statements for Qualitative Features
TL;DR: In this article, the authors derive multiscale statistics for deconvolution in order to detect qualitative features of the unknown density of a density in the context of local monotonicity on all scales simultaneously.