We propose using the integrated periodogram to classify time series. The method assigns a new element to the group minimizing the distance from the integratedperiodogram of the element to the group mean of integrated periodograms. Localcomputation of these periodograms allows the application of the approach to non--stationary time series. Since the integrated periodograms are functional data, weapply depth-based techniques to make the classification robust. The methodprovides small error rates with both simulated and real data, and shows goodcomputational behaviour.
展开▼
