A new method for generating random correlation matrices

B-Tier
Journal: The Econometrics Journal
Year: 2024
Volume: 27
Issue: 2
Pages: 188-212

Score contribution per author:

0.670 = (α=2.01 / 3 authors) × 1.0x B-tier

α: calibrated so average coauthorship-adjusted count equals average raw count

Abstract

SummaryWe propose a new method for generating random correlation matrices that makes it simple to control both location and dispersion. The method is based on a vector parameterization, , which maps any distribution onto a distribution on the space of nonsingularcorrelation matrices. Correlation matrices with certain properties, such as being well-conditioned, having block structures, and having strictly positive elements, are simple to generate. We compare the new method with existing methods.

Technical Details

RePEc Handle
repec:oup:emjrnl:v:27:y:2024:i:2:p:188-212.
Journal Field
Econometrics
Author Count
3
Added to Database
2026-01-24