Calibrated precision matrix estimation for high-dimensional elliptical distributions

Tuo Zhao; Han Liu

doi:10.1109/TIT.2014.2360980

Calibrated precision matrix estimation for high-dimensional elliptical distributions

Tuo Zhao, Han Liu

Computer Science

Research output: Contribution to journal › Article › peer-review

17 Scopus citations

Abstract

We propose a semiparametric method for estimating a precision matrix of high-dimensional elliptical distributions. Unlike most existing methods, our method naturally handles heavy tailness and conducts parameter estimation under a calibration framework, thus achieves improved theoretical rates of convergence and finite sample performance on heavy-tail applications. We further demonstrate the performance of the proposed method using thorough numerical experiments.

Original language	English (US)
Article number	6913534
Pages (from-to)	7884-7887
Number of pages	4
Journal	IEEE Transactions on Information Theory
Volume	60
Issue number	12
DOIs	https://doi.org/10.1109/TIT.2014.2360980
State	Published - Dec 1 2014

Keywords

Calibrated Estimation
Elliptical Distribution
Heavy-tailness
Precision Matrix
Semiparametric Model

ASJC Scopus subject areas

Information Systems
Computer Science Applications
Library and Information Sciences

Access to Document

10.1109/TIT.2014.2360980

Cite this

@article{47f5003c8ab348da8013f1b6a8417d3f,

title = "Calibrated precision matrix estimation for high-dimensional elliptical distributions",

abstract = "We propose a semiparametric method for estimating a precision matrix of high-dimensional elliptical distributions. Unlike most existing methods, our method naturally handles heavy tailness and conducts parameter estimation under a calibration framework, thus achieves improved theoretical rates of convergence and finite sample performance on heavy-tail applications. We further demonstrate the performance of the proposed method using thorough numerical experiments.",

keywords = "Calibrated Estimation, Elliptical Distribution, Heavy-tailness, Precision Matrix, Semiparametric Model",

author = "Tuo Zhao and Han Liu",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.",

year = "2014",

month = dec,

day = "1",

doi = "10.1109/TIT.2014.2360980",

language = "English (US)",

volume = "60",

pages = "7884--7887",

journal = "IEEE Transactions on Information Theory",

issn = "0018-9448",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "12",

}

TY - JOUR

T1 - Calibrated precision matrix estimation for high-dimensional elliptical distributions

AU - Zhao, Tuo

AU - Liu, Han

PY - 2014/12/1

Y1 - 2014/12/1

N2 - We propose a semiparametric method for estimating a precision matrix of high-dimensional elliptical distributions. Unlike most existing methods, our method naturally handles heavy tailness and conducts parameter estimation under a calibration framework, thus achieves improved theoretical rates of convergence and finite sample performance on heavy-tail applications. We further demonstrate the performance of the proposed method using thorough numerical experiments.

AB - We propose a semiparametric method for estimating a precision matrix of high-dimensional elliptical distributions. Unlike most existing methods, our method naturally handles heavy tailness and conducts parameter estimation under a calibration framework, thus achieves improved theoretical rates of convergence and finite sample performance on heavy-tail applications. We further demonstrate the performance of the proposed method using thorough numerical experiments.

KW - Calibrated Estimation

KW - Elliptical Distribution

KW - Heavy-tailness

KW - Precision Matrix

KW - Semiparametric Model

UR - http://www.scopus.com/inward/record.url?scp=84912119612&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84912119612&partnerID=8YFLogxK

U2 - 10.1109/TIT.2014.2360980

DO - 10.1109/TIT.2014.2360980

M3 - Article

C2 - 25632164

AN - SCOPUS:84912119612

SN - 0018-9448

VL - 60

SP - 7884

EP - 7887

JO - IEEE Transactions on Information Theory

JF - IEEE Transactions on Information Theory

IS - 12

M1 - 6913534

ER -

Calibrated precision matrix estimation for high-dimensional elliptical distributions

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this