Robustness of a Noise Filtering Procedure for Correlation Matrices : B. Filtering of Deliberately Introduced Noise from 'True' Correlation Matrices
Demonstration that our noise filtering procedure is extremely robust on the basis of the following experiment. The noise filtering procedure was applied first to an empirical correlation matrix and, second, to the matrix built from the same time series deliberately contaminated with noise. The final, noise filtered correlation matrices, were practically the same in both cases (by simple visual comparison). This conclusion was confirmed by entropy measures as well as by Euclidean measures that gave average distance between off-diagonal elements of correlation matrices obtained as a result of noise filtering of empirical correlation matrix with and without deliberate noise contamination of the underlying time series
Year of publication: |
2014
|
---|---|
Authors: | Izmailov, Alexander |
Other Persons: | Shay, Brian (contributor) |
Publisher: |
[2014]: [S.l.] : SSRN |
Subject: | Korrelation | Correlation | Zeitreihenanalyse | Time series analysis | Schätztheorie | Estimation theory | Zustandsraummodell | State space model |
Saved in:
freely available
Extent: | 1 Online-Ressource (12 p) |
---|---|
Type of publication: | Book / Working Paper |
Language: | English |
Notes: | Nach Informationen von SSRN wurde die ursprüngliche Fassung des Dokuments June 3, 2013 erstellt |
Other identifiers: | 10.2139/ssrn.2378974 [DOI] |
Classification: | C1 - Econometric and Statistical Methods: General ; C4 - Econometric and Statistical Methods: Special Topics ; C6 - Mathematical Methods and Programming ; G1 - General Financial Markets |
Source: | ECONIS - Online Catalogue of the ZBW |
Persistent link: https://www.econbiz.de/10013060875