Journal
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
Volume 389, Issue 13, Pages 2566-2579Publisher
ELSEVIER
DOI: 10.1016/j.physa.2010.02.026
Keywords
Random matrix; Financial covariances; Local statistics; Power law; Tracy-Widom; Wigner's surmise
Categories
Funding
- European Community [MRTN-CT-2004-005616]
Ask authors/readers for more resources
We investigate whether quantities such as the global spectral density or individual eigenvalues of financial covariance matrices can be best modelled by standard random matrix theory or rather by its generalisations displaying power-law tails. In order to generate individual eigenvalue distributions a chopping procedure is devised, which produces a statistical ensemble of asset-price covariances from a single instance of financial data sets. Local results for the smallest eigenvalue and individual spacings are very stable upon reshuffling the time windows and assets. They are in good agreement with the universal Tracy-Widom distribution and Wigner surmise, respectively. This suggests a strong degree of robustness especially in the low-lying sector of the spectra, most relevant for portfolio selections. Conversely, the global spectral density of a single covariance matrix as well as the average over all unfolded nearest-neighbour spacing distributions deviate from standard Gaussian random matrix predictions. The data are in fair agreement with a recently introduced generalised random matrix model, with correlations showing a power-law decay. (C) 2010 Elsevier B.V. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available