On the utility of asymptotic bandwidth selectors for spatially adaptive kernel density estimation

Tilman M. Davies; Claire R. Flynn; Martin L. Hazelton

doi:10.1016/j.spl.2018.02.067

Back

On the utility of asymptotic bandwidth selectors for spatially adaptive kernel density estimation

Journal article

Peer reviewed

On the utility of asymptotic bandwidth selectors for spatially adaptive kernel density estimation

Tilman M. Davies, Claire R. Flynn and Martin L. Hazelton

Statistics & probability letters, Vol.138, pp.75-81

07/2018

DOI: https://doi.org/10.1016/j.spl.2018.02.067

Abstract

AMISE

Global bandwidth

Kernel smoothing

Normal scale rule

Spatial point process

Implementation of the spatially adaptive kernel estimator relies on choice of a ‘global bandwidth’. We derive the closed-form asymptotic bias for this estimator with the aim of developing relevant selectors, and note non-uniform convergence hinders its usability for this purpose.

Metrics

1 Record Views

Details

Record Identifier: 9926515263701891
Title: On the utility of asymptotic bandwidth selectors for spatially adaptive kernel density estimation
Creators: Tilman M. Davies
Claire R. Flynn
Martin L. Hazelton
Publication Details: Statistics & probability letters, Vol.138, pp.75-81
Academic Unit: Mathematics and Statistics
Publisher: Elsevier B.V
Date published ; e-published: 07/2018
Language: English
Resource Type: Journal article

On the utility of asymptotic bandwidth selectors for spatially adaptive kernel density estimation

Abstract

Related links

Metrics

Details

Usage Policy