A Wild Bootstrap Procedure for the Identification of Optimal Groups in Singular Spectrum Analysis

A key step in separating signal from noise using Singular Spectrum Analysis (SSA) is grouping, which is often done subjectively. In this article a method which enables the identification of statistically significant groups for the grouping step in SSA is presented. The proposed procedure provides a more objective and reliable approach for separating noise from the main signal in SSA. We utilize the w- correlation and test if it close or equal to zero. A wild bootstrap approach is used to determine the distribution of the w-correlation. To identify an ideal number of groupings which leads to almost perfect separation of the noise and signal, a given number of groups are tested, necessitating accounting for multiplicity. The effectiveness of our method in identifying the best group is demonstrated through a simulation study, furthermore, we have applied the approach to real world data in the context of neuroimaging. This research provides a valuable contribution to the field of SSA and offers important insights into the statistical properties of the w-correlation distribution. The results obtained from the simulation studies and analysis of real-world data demonstrate the effectiveness of the proposed approach in identifying the best groupings for SSA.