Title
Feature significance for multivariate kernel density estimation
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.
Year
DOI
Venue
2008
10.1016/j.csda.2008.02.035
Computational Statistics & Data Analysis
Keywords
Field
DocType
kernel density derivative estimator,modal regions.,modal region,hypothesis test,multivariate kernel density estimation,gradient estimators,d-dimensional data,kernel curvature estimator,feature signiflcance testing,three-dimensional data,curvature estimators,feature significance,kernel gradient estimator,density derivatives,scale space,kernel density estimate,density estimation,statistical significance,kernel density,three dimensional
Econometrics,Density estimation,Multivariate kernel density estimation,Kernel embedding of distributions,Kernel principal component analysis,Kernel method,Statistics,Variable kernel density estimation,Mathematics,Kernel (statistics),Kernel density estimation
Journal
Volume
Issue
ISSN
52
9
Computational Statistics and Data Analysis
Citations 
PageRank 
References 
17
2.60
0
Authors
4
Name
Order
Citations
PageRank
Tarn Duong1224.53
Arianna Cowling2172.60
Inge Koch3376.58
M. P. Wand45110.35