Cover Image

The conditioning of the maximum entropy covariance matrix and its inverse

The maximum entropy covariance matrix is positive definite even when the number of variables p exceeds the sample size n. However, the inverse of this matrix can have stability problems when p is close to n, although these problems tend to disappear as p increases beyond n. We analyze such problems using the variance of the latent roots in a particular metric as a condition number.
Year of publication:
1982
Authors: Conway, Delores ; Theil, Henri
Published in:
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 1.1982, 2, p. 103-106
Publisher: Elsevier
Keywords: Condition numbers covariance matrix estimation entropy latent roots ridge matrices
Type of publication: Article
Source:
Persistent link: https://www.econbiz.de/10005137982
Similar items by person