Improved variance estimation along sample eigenvectors

A.J. Hendrikse; Raymond N.J. Veldhuis; Lieuwe Jan Spreeuwers

Improved variance estimation along sample eigenvectors

A.J. Hendrikse, Raymond N.J. Veldhuis, Lieuwe Jan Spreeuwers

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic

41 Downloads (Pure)

Abstract

Second order statistics estimates in the form of sample eigenvalues and sample eigenvectors give a sub optimal description of the population density. So far only attempts have been made to reduce the bias in the sample eigenvalues. However, because the sample eigenvectors differ from the population eigenvectors as well, the population eigenvalues are biased estimates of the variances along the sample eigenvectors. Therefore correction of the sample eigenvalues towards the population eigenvalues is not sufficient. The experiments in this paper show that replacing the sample eigenvalues with the variances along the sample eigenvectors often results in better estimates of the population density than replacing the sample eigenvalues with the population eigenvalues.

Original language	Undefined
Title of host publication	Proceedings of the 30th Symposium on Information Theory in the Benelux
Place of Publication	Eindhoven
Publisher	Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)
Pages	25-32
Number of pages	8
ISBN (Print)	978-90-386-1852-4
Publication status	Published - May 2009
Event	30th WIC Symposium on Information Theory in the Benelux 2009 - Eindhoven, Netherlands Duration: 28 May 2009 → 29 May 2009 Conference number: 30

Publication series

Name
Publisher	Werkgemeenschap voor Informatie- en Communicatietheorie

Conference

Conference	30th WIC Symposium on Information Theory in the Benelux 2009
Country/Territory	Netherlands
City	Eindhoven
Period	28/05/09 → 29/05/09

Keywords

IR-70158
METIS-265748
Density estimation
EWI-15685
eigenvalue estimation
SCS-Safety
variance correction

Access to Document

wic2009_Hendrikse_WIC30-17

Cite this

@inproceedings{7ce02c226107441fa7e0a36a83480ccc,

title = "Improved variance estimation along sample eigenvectors",

abstract = "Second order statistics estimates in the form of sample eigenvalues and sample eigenvectors give a sub optimal description of the population density. So far only attempts have been made to reduce the bias in the sample eigenvalues. However, because the sample eigenvectors differ from the population eigenvectors as well, the population eigenvalues are biased estimates of the variances along the sample eigenvectors. Therefore correction of the sample eigenvalues towards the population eigenvalues is not sufficient. The experiments in this paper show that replacing the sample eigenvalues with the variances along the sample eigenvectors often results in better estimates of the population density than replacing the sample eigenvalues with the population eigenvalues.",

keywords = "IR-70158, METIS-265748, Density estimation, EWI-15685, eigenvalue estimation, SCS-Safety, variance correction",

author = "A.J. Hendrikse and Veldhuis, {Raymond N.J.} and Spreeuwers, {Lieuwe Jan}",

year = "2009",

month = may,

language = "Undefined",

isbn = "978-90-386-1852-4",

publisher = "Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)",

pages = "25--32",

booktitle = "Proceedings of the 30th Symposium on Information Theory in the Benelux",

address = "Netherlands",

note = "30th WIC Symposium on Information Theory in the Benelux 2009 ; Conference date: 28-05-2009 Through 29-05-2009",

}

Improved variance estimation along sample eigenvectors. / Hendrikse, A.J.; Veldhuis, Raymond N.J.; Spreeuwers, Lieuwe Jan.
Proceedings of the 30th Symposium on Information Theory in the Benelux. Eindhoven: Werkgemeenschap voor Informatie- en Communicatietheorie (WIC), 2009. p. 25-32.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic

TY - GEN

T1 - Improved variance estimation along sample eigenvectors

AU - Hendrikse, A.J.

AU - Veldhuis, Raymond N.J.

AU - Spreeuwers, Lieuwe Jan

N1 - Conference code: 30

PY - 2009/5

Y1 - 2009/5

N2 - Second order statistics estimates in the form of sample eigenvalues and sample eigenvectors give a sub optimal description of the population density. So far only attempts have been made to reduce the bias in the sample eigenvalues. However, because the sample eigenvectors differ from the population eigenvectors as well, the population eigenvalues are biased estimates of the variances along the sample eigenvectors. Therefore correction of the sample eigenvalues towards the population eigenvalues is not sufficient. The experiments in this paper show that replacing the sample eigenvalues with the variances along the sample eigenvectors often results in better estimates of the population density than replacing the sample eigenvalues with the population eigenvalues.

AB - Second order statistics estimates in the form of sample eigenvalues and sample eigenvectors give a sub optimal description of the population density. So far only attempts have been made to reduce the bias in the sample eigenvalues. However, because the sample eigenvectors differ from the population eigenvectors as well, the population eigenvalues are biased estimates of the variances along the sample eigenvectors. Therefore correction of the sample eigenvalues towards the population eigenvalues is not sufficient. The experiments in this paper show that replacing the sample eigenvalues with the variances along the sample eigenvectors often results in better estimates of the population density than replacing the sample eigenvalues with the population eigenvalues.

KW - IR-70158

KW - METIS-265748

KW - Density estimation

KW - EWI-15685

KW - eigenvalue estimation

KW - SCS-Safety

KW - variance correction

M3 - Conference contribution

SN - 978-90-386-1852-4

SP - 25

EP - 32

BT - Proceedings of the 30th Symposium on Information Theory in the Benelux

PB - Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)

CY - Eindhoven

T2 - 30th WIC Symposium on Information Theory in the Benelux 2009

Y2 - 28 May 2009 through 29 May 2009

ER -