White Rose University Consortium logo
University of Leeds logo University of Sheffield logo York University logo

Automatic bandwidth selection for circular density estimation

Taylor, C.C. (2008) Automatic bandwidth selection for circular density estimation. Computational Statistics and Data Analysis, 52 (7). pp. 3493-3500. ISSN 0167-9473

Available under licence : See the attached licence file.

Download (197Kb)


Given angular data θ1,…,θn[0,2π) a common objective is to estimate the density. In case that a kernel estimator is used, bandwidth selection is crucial to the performance. A “plug-in rule” for the bandwidth, which is based on the concentration of a reference density, namely, the von Mises distribution is obtained. It is seen that this is equivalent to the usual Euclidean plug-in rule in the case where the concentration becomes large. In case that the concentration parameter is unknown, alternative methods are explored which are intended to be robust to departures from the reference density. Simulations indicate that “wrapped estimators” can perform well in this context. The methods are applied to a real bivariate dataset concerning protein structure.

Item Type: Article
Copyright, Publisher and Additional Information: © 2008 Elsevier B.V. This is an author produced version of a paper published in Computational Statistics and Data Analysis . Uploaded in accordance with the publisher's self-archiving policy.
Keywords: angle data, kernel density estimators, von Mises distribution, Ramachandran plot, Smoothing parameter selection
Institution: The University of Leeds
Academic Units: The University of Leeds > Faculty of Maths and Physical Sciences (Leeds) > School of Mathematics (Leeds) > Statistics (Leeds)
Depositing User: Sherpa Assistant
Date Deposited: 16 May 2008 08:48
Last Modified: 08 Feb 2013 17:05
Published Version: http://dx.doi.org/10.1016/j.csda.2007.11.003
Status: Published
Publisher: Elsevier Science B.V
Identification Number: 10.1016/j.csda.2007.11.003
URI: http://eprints.whiterose.ac.uk/id/eprint/3793

Actions (repository staff only: login required)