SHORT-TERM FORECASTING OF TIME SERIES BASED ON EVOLUTIONAL ALGORITHMS
M. Danilov · Applied Mathematics and Control Sciences · 2019
The paper proposes a method of short-term forecasting based on the identification of the spanning algebraic sequence c using evolutionary algorithms. This method is especially effective in the case of short time series. When there is not enough data for training standard models, the developed method still allows to extract the maximum available information on the process behavior in order to extrapolate it to future points in time.