Generating random correlation matrices by the simple rejection method: Why it does not work.

Walter Böhm; Kurt Hornik

doi:10.1016/j.spl.2013.12.012

Generating random correlation matrices by the simple rejection method: Why it does not work.

Walter Böhm, Kurt Hornik

Statistics and Mathematics

Publikation: Wissenschaftliche Fachzeitschrift › Originalbeitrag in Fachzeitschrift › Begutachtung

Abstract

We derive exact and asymptotic formulas for the probability that a symmetric n×n matrix with unit diagonal and upper diagonal elements i.i.d. uniform on (−1,1) is positive definite (and thus a “random correlation matrix”): this is almost never the case for n≥6.

Originalsprache	Englisch
Seiten (von - bis)	27 - 30
Fachzeitschrift	Statistics and Probability Letters
Jahrgang	87
DOIs	https://doi.org/10.1016/j.spl.2013.12.012
Publikationsstatus	Veröffentlicht - 2014

Zugriff auf Dokument

10.1016/j.spl.2013.12.012Lizenz: Nicht angegeben

Verfügbarkeit prüfen

Katalog der Universitätsbibliothek

Andere Dateien und Links

http://www.sciencedirect.com/science/article/pii/S0167715213004112

Dieses zitieren

@article{29150ed3aca446228aa2f5451eee48d2,

title = "Generating random correlation matrices by the simple rejection method: Why it does not work.",

abstract = "We derive exact and asymptotic formulas for the probability that a symmetric n×n matrix with unit diagonal and upper diagonal elements i.i.d. uniform on (−1,1) is positive definite (and thus a “random correlation matrix”): this is almost never the case for n≥6.",

author = "Walter B{\"o}hm and Kurt Hornik",

year = "2014",

doi = "10.1016/j.spl.2013.12.012",

language = "English",

volume = "87",

pages = "27 -- 30",

journal = "Statistics and Probability Letters",

issn = "0167-7152",

publisher = "Elsevier",

}

TY - JOUR

T1 - Generating random correlation matrices by the simple rejection method: Why it does not work.

AU - Böhm, Walter

AU - Hornik, Kurt

PY - 2014

Y1 - 2014

N2 - We derive exact and asymptotic formulas for the probability that a symmetric n×n matrix with unit diagonal and upper diagonal elements i.i.d. uniform on (−1,1) is positive definite (and thus a “random correlation matrix”): this is almost never the case for n≥6.

AB - We derive exact and asymptotic formulas for the probability that a symmetric n×n matrix with unit diagonal and upper diagonal elements i.i.d. uniform on (−1,1) is positive definite (and thus a “random correlation matrix”): this is almost never the case for n≥6.

UR - http://www.sciencedirect.com/science/article/pii/S0167715213004112

U2 - 10.1016/j.spl.2013.12.012

DO - 10.1016/j.spl.2013.12.012

M3 - Journal article

VL - 87

SP - 27

EP - 30

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

SN - 0167-7152

ER -