CiSSA: Circulant SSA under Matlab

Circulant Singular Spectrum Analysis is a procedure for signal processing

A set of MATLAB functions is presented for signal extraction by Circulant Singular Spectrum Analysis (SSA), a procedure proposed in Bógalo et al (2021). Circulant SSA is a new variant of SSA that allows to extract the signal associated to any frequency speci ed beforehand. This is a novelty when compared with other SSA procedures that need to identify ex-post the frequencies associated to the extracted signals. Circulant SSA offers a solution based on circulant matrices and presents an automated procedure to compute the optimal reconstruction. This solution can be applied to the non-stationary series, and in the non-linear case. The procedure is non-parametric and it has been automated so as the final user only needs to give the data and specify the window length. It works well for economic time series as well as for series coming from the physical world with both modulated amplitude and frequency (AM-FM signals).