MCiSSA: Multivariate Circulant SSA under Matlab

Multivariate Circulant Singular Spectrum Analysis is a procedure for signal processing of multivariate time series

A set of MATLAB functions is presented for signals extraction by Multivariate Circulant Singular Spectrum Analysis (MCiSSA), a procedure proposed in Bógalo et al (2024). Multivariate Circulant SSA is a new variant of Multivariate SSA that allows to extract the signals associated to any frequency specied beforehand. The key for the self-identification of the frequencies is the use of block circulant matrices of second moments, instead of the usual variance-covariance matrix, that allows matching eigenvalues with frequencies. Additionally, MCiSSA performs a double diagonalization of that block circulant matrix that uncovers cross section relations per frequency.