Artificial Immune Clustering Algorithm to Forecasting Seasonal Time Series

This paper concentrates on the forecasting time series with multiple seasonal periods using new immune inspired method. Proposed model includes two populations of immune memory cells – antibodies, which recognize patterns of the time series sequences represented by antigens. The empirical probabilities, that the pattern of forecasted sequence is detected by the jth antibody from the first population while the corresponding pattern of input sequence is detected by the ith antibody from the second population, are computed and applied to the forecast construction. The suitability of the proposed approach is illustrated through an application to electrical load forecasting.