scientific article; zbMATH DE number 4032808
Let \(F_ n(x)\) be an empirical distribution function based on independent observations of the random variable \(\xi\) with differentiable density f(x). The author considers the estimates for derivatives of f(x) in the point \(x_ 0\in R:\) \[ f'_ n(x_ 0)=h_ n^{- 2}\int^{\infty}_{-\infty}w((x-x_ 0)/h_ n)dF_ n(x),\quad f_ n(x_ 0)=2h_ n^{-3}\int^{\infty}_{-\infty}w((x-x_ 0)/h_ n)dF_ n(x), \] where \(h_ n\) and w(x) satisfy some conditions. Based on \(f_ n'(x_ 0)\) and \(f_ n(x_ 0)\) statistical estimates for extreme points and inflection points of the density f(x) are given and two theorems on the rate of convergence of these estimates are proved.
- Higher order estimation at Lebesgue points
- Asymptotic distributions of estimators of the derivative of a density and the score function
- On the impossibility of estimating densities in the extreme tail
- Nonparametric estimation of the ratios of derivatives of a multivariate distribution density from dependent observations
- scientific article; zbMATH DE number 3862234 (Why is no real title available?)
- scientific article; zbMATH DE number 1163788 (Why is no real title available?)
- scientific article; zbMATH DE number 2151912 (Why is no real title available?)
- On Qualitative Smoothness of Kernel Density Estimates1
- On the oscillation of the expected number of extreme points of a random set
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3773071)