A New Evolutionary Hybrid Random Forest Model for SPEI Forecasting

State-of-the-art random forest (RF) models have been documented as versatile tools to solve regression and classification problems in hydrology. They can model stochastic time series by bagging different decision trees. This article introduces a new hybrid RF that increases the forecasting accuracy of RF-based models. The model, called GARF, is attained integrating genetic algorithm (GA) (RF), which trees are bagged. We applied GARF forecast multitemporal drought index (SPEI-3 SPEI-6) at two meteorology stations (Beypazari Nallihan) Ankara, Turkey. compared associated results with classic RF, standalone extreme learning machine (ELM), ELM optimized Bat (Bat-ELM) verify accuracy. performance assessment was performed using graphical statistical analysis. demonstrated outperformed benchmark achieved least error quantitative for prediction both SPEI-3 SPEI-6, particularly testing period. this study showed improve technique up 30% 40% Beypazari Nallihan stations, respectively.