Table 6.

Forecasting performance indices of models for combination 2.

Model	NSE	d	KGE	MAE	MBE	RE	RMSE	PCC	R2
	Training Data set (N = 1284)
Linear Regression	0.638	0.855	0.607	0.598	−0.014	−1.400%	0.830	0.815	0.664
LR-SVM	0.660	0.874	0.665	0.548	−0.012	−1.200%	0.805	0.819	0.671
LR-RSS	0.849	0.955	0.840	0.290	0.009	0.900%	0.536	0.924	0.854
LR-REPTree	0.993	0.998	0.995	0.042	0.006	0.600%	0.119	0.996	0.992
LR-M5P	0.964	0.990	0.933	0.137	0.002	0.200%	0.261	0.983	0.966
	Testing Data set (N = 550)
Linear Regression	0.705	0.887	0.635	0.563	−0.162	−16.200%	0.737	0.875	0.766
LR-SVM	0.734	0.906	0.700	0.503	−0.136	−13.600%	0.700	0.875	0.766
LR-RSS	0.865	0.961	0.864	0.278	−0.015	−8.400%	0.498	0.932	0.869
LR-REPTree	0.941	0.984	0.923	0.143	−0.089	−0.900%	0.331	0.973	0.947
LR-M5P	0.950	0.986	0.892	0.181	−0.084	−1.500%	0.304	0.980	0.960