Algorithm for the Detection of Changes of the Correlation Structure in Multivariate Time Series
Keywords:Covariance matrix, entropy, principal component analysis
AbstractIn this research, an improved algorithm for the detection of changes of the correlation structure in multivariate time series is proposed. The starting point of the technique is a covariance matrix whose entries are the largest entries of a cross-covariance matrix which is composed of all pairs of the time series reconstructed to an M-dimensional phase space. Principal component analysis is performed on this maximized cross-covariance matrix, and the overall degree of synchronization among multiple-channel signals is defined, by synchronization index, as the Shannon entropy of the eigenvalue spectrum. Throughout the experiment, the effectiveness of the proposed algorithm is validated with simulated data – a network of time series generated by autoregressive models and a network of coupled chaotic Roessler oscillators.
How to Cite
The copyright for the paper in this journal is retained by the author(s) with the first publication right granted to the journal. The authors agree to the Creative Commons Attribution 4.0 (CC BY 4.0) agreement under which the paper in the Journal is licensed.
By virtue of their appearance in this open access journal, papers are free to use with proper attribution in educational and other non-commercial settings with an acknowledgement of the initial publication in the journal.