A Canonical Representation of Block Matrices with Applications to Covariance and Correlation Matrices

12/04/2020
by   Ilya Archakov, et al.
0

We obtain a canonical representation for block matrices. The representation facilitates simple computation of the determinant, the matrix inverse, and other powers of a block matrix, as well as the matrix logarithm and the matrix exponential. These results are particularly useful for block covariance and block correlation matrices, where evaluation of the Gaussian log-likelihood and estimation are greatly simplified. We illustrate this with an empirical application using a large panel of daily asset returns. Moreover, the representation paves new ways to regularizing large covariance/correlation matrices and to test block structures in matrices.

READ FULL TEXT

page 20

page 21

research
08/11/2023

Characterizing Correlation Matrices that Admit a Clustered Factor Representation

The Clustered Factor (CF) model induces a block structure on the correla...
research
04/17/2023

A New Representation of Uniform-Block Matrix and Applications

A covariance matrix with a special pattern (e.g., sparsity or block stru...
research
12/04/2020

A New Parametrization of Correlation Matrices

We introduce a novel parametrization of the correlation matrix. The repa...
research
12/07/2020

Likelihood Geometry of Correlation Models

Correlation matrices are standardized covariance matrices. They form an ...
research
06/19/2017

Detection of Block-Exchangeable Structure in Large-Scale Correlation Matrices

Correlation matrices are omnipresent in multivariate data analysis. When...
research
02/07/2018

Group kernels for Gaussian process metamodels with categorical inputs

Gaussian processes (GP) are widely used as a metamodel for emulating tim...
research
06/26/2018

Estimation of large block covariance matrices: Application to the analysis of gene expression data

Motivated by an application in molecular biology, we propose a novel, ef...

Please sign up or login with your details

Forgot password? Click here to reset