Feature significance for multivariate kernel density estimation

Duong, T; Cowling, A; Koch, I; Wand, MP

Feature significance for multivariate kernel density estimation

Duong, T Cowling, A Koch, I Wand, MP

Permalink

Publication Type:: Journal Article
Citation:: Computational Statistics and Data Analysis, 2008, 52 (9), pp. 4225 - 4242
Issue Date:: 2008-05-15

Closed Access

	Filename	Description	Size
	2010000264OK.pdf		1.34 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Duong, T	en_US
dc.contributor.author	Cowling, A	en_US
dc.contributor.author	Koch, I	en_US
dc.contributor.author	Wand, MP https://orcid.org/0000-0003-2555-896X	en_US
dc.date.issued	2008-05-15	en_US
dc.identifier.citation	Computational Statistics and Data Analysis, 2008, 52 (9), pp. 4225 - 4242	en_US
dc.identifier.issn	0167-9473	en_US
dc.identifier.uri	http://hdl.handle.net/10453/13472
dc.description.abstract	Multivariate kernel density estimation provides information about structure in data. Feature significance is a technique for deciding whether features-such as local extrema-are statistically significant. This paper proposes a framework for feature significance in d-dimensional data which combines kernel density derivative estimators and hypothesis tests for modal regions. For the gradient and curvature estimators distributional properties are given, and pointwise test statistics are derived. The hypothesis tests extend the two-dimensional feature significance ideas of Godtliebsen et al. [Godtliebsen, F., Marron, J.S., Chaudhuri, P., 2002. Significance in scale space for bivariate density estimation. Journal of Computational and Graphical Statistics 11, 1-21]. The theoretical framework is complemented by novel visualization for three-dimensional data. Applications to real data sets show that tests based on the kernel curvature estimators perform well in identifying modal regions. These results can be enhanced by corresponding tests with kernel gradient estimators. © 2008 Elsevier Ltd. All rights reserved.	en_US
dc.relation.ispartof	Computational Statistics and Data Analysis	en_US
dc.relation.isbasedon	10.1016/j.csda.2008.02.035	en_US
dc.subject.classification	Statistics & Probability	en_US
dc.title	Feature significance for multivariate kernel density estimation	en_US
dc.type	Journal Article
utslib.citation.volume	9	en_US
utslib.citation.volume	52	en_US
utslib.for	0104 Statistics	en_US
utslib.for	0802 Computation Theory and Mathematics	en_US
utslib.for	1403 Econometrics	en_US
dc.location.activity	Taipei, TAIWAN
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Science
pubs.organisational-group	/University of Technology Sydney/Faculty of Science/School of Mathematical and Physical Sciences
utslib.copyright.status	closed_access
pubs.issue	9	en_US
pubs.publication-status	Published	en_US
pubs.volume	52	en_US

Abstract:

Multivariate kernel density estimation provides information about structure in data. Feature significance is a technique for deciding whether features-such as local extrema-are statistically significant. This paper proposes a framework for feature significance in d-dimensional data which combines kernel density derivative estimators and hypothesis tests for modal regions. For the gradient and curvature estimators distributional properties are given, and pointwise test statistics are derived. The hypothesis tests extend the two-dimensional feature significance ideas of Godtliebsen et al. [Godtliebsen, F., Marron, J.S., Chaudhuri, P., 2002. Significance in scale space for bivariate density estimation. Journal of Computational and Graphical Statistics 11, 1-21]. The theoretical framework is complemented by novel visualization for three-dimensional data. Applications to real data sets show that tests based on the kernel curvature estimators perform well in identifying modal regions. These results can be enhanced by corresponding tests with kernel gradient estimators. © 2008 Elsevier Ltd. All rights reserved.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/13472