Hydroclimatic Data Prediction using a New Ensemble Group Method of Data Handling Coupled with Artificial Bee Colony Algorithm

Linear regression is widely used in flood quantile study that consists of meteorological and physiographical variables. However, linear does not capture the complex nonlinear relationship between predictor target It rare to find a hydrological application using group method data handling (GMDH) model, artificial bee colony (ABC) algorithm, ensemble technique, precisely predicting ungauged sites. GMDH model known be an effective complying with relationship. Therefore, this paper, we enhance by implementing ABC algorithm optimize parameter partial description some transfer functions, namely polynomial, radial basis, sigmoid hyperbolic tangent function. Then, averaging combines output from those various functions becomes new coupled (EGMDH-ABC) model. The results show significantly improves prediction performance EGMDH-ABC satisfies nonlinearity produce better estimation. Also, it provides more robust, accurate, efficient results.