A Meta-Genetic Algorithm for Time Series Forecasting
Paulo Cortez, Miguel Rocha, José Neves · Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT) · 2001
Alternative approaches for Time Series Forecasting (TSF) emerged from the Artificial Intelligence arena, where optimization algorithms inspired on natural selection processes, such as Genetic Algorithms (GAs) are popular. The present work reports on a two-level architecture, where a (meta-level) binary GA will search for the best TSF model, being the parameters optimized by a (low-level) GA, which encodes real values. The machine's performance of this approach was compared with conventional forecasting methods, exhibiting good results,specially when trended and nonlinear series are considered.