Correlation matrices with average constraints

Jan Tuitman; Steven Vanduffel; Jing Yao

doi:10.1016/j.spl.2020.108868

Correlation matrices with average constraints

Jan Tuitman, Steven Vanduffel, Jing Yao

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

4 Citations (Scopus)

112 Downloads (Pure)

Abstract

We develop an algorithm that makes it possible to generate all correlation matrices satisfying a constraint on their average value. We extend the results to the case of multiple constraints. These results can be used to assess the extent to which methodologies driven by correlation matrices are robust to misspecification thereof.

Original language	English
Article number	108868
Journal	Statistics and Probability Letters
Volume	165
Early online date	7 Jul 2020
DOIs	https://doi.org/10.1016/j.spl.2020.108868
Publication status	Published - Oct 2020

Access to Document

10.1016/j.spl.2020.108868

Download pdf
© 2020 Elsevier B.V.
Accepted author manuscript, 290 KBLicence: CC BY-NC-ND

Cite this

@article{b48b04f64e404de683951f39c3cb4d93,

title = "Correlation matrices with average constraints",

abstract = "We develop an algorithm that makes it possible to generate all correlation matrices satisfying a constraint on their average value. We extend the results to the case of multiple constraints. These results can be used to assess the extent to which methodologies driven by correlation matrices are robust to misspecification thereof.",

author = "Jan Tuitman and Steven Vanduffel and Jing Yao",

year = "2020",

month = oct,

doi = "10.1016/j.spl.2020.108868",

language = "English",

volume = "165",

journal = "Statistics and Probability Letters",

issn = "0167-7152",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Correlation matrices with average constraints

AU - Tuitman, Jan

AU - Vanduffel, Steven

AU - Yao, Jing

PY - 2020/10

Y1 - 2020/10

N2 - We develop an algorithm that makes it possible to generate all correlation matrices satisfying a constraint on their average value. We extend the results to the case of multiple constraints. These results can be used to assess the extent to which methodologies driven by correlation matrices are robust to misspecification thereof.

AB - We develop an algorithm that makes it possible to generate all correlation matrices satisfying a constraint on their average value. We extend the results to the case of multiple constraints. These results can be used to assess the extent to which methodologies driven by correlation matrices are robust to misspecification thereof.

U2 - 10.1016/j.spl.2020.108868

DO - 10.1016/j.spl.2020.108868

M3 - Article

SN - 0167-7152

VL - 165

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

M1 - 108868

ER -

Correlation matrices with average constraints

Abstract

Access to Document

Fingerprint

Cite this