Table 5.

Forecasting performance indices of models for combination 1.

Model	NSE	d	KGE	MAE	MBE	RE	RMSE	PCC	R2
	Training Data set (N = 1284)
Linear Regression	0.700	0.899	0.743	0.459	−0.009	−0.900%	0.757	0.837	0.701
LR-SVM	0.716	0.907	0.758	0.451	−0.009	−0.900%	0.736	0.847	0.717
LR-RSS	0.936	0.982	0.905	0.177	0.002	0.200%	0.348	0.969	0.939
LR-REPTree	0.995	0.999	0.996	0.033	−0.001	−0.100%	0.094	0.998	0.996
LR-M5P	0.771	0.922	0.740	0.439	−0.016	−1.600%	0.660	0.887	0.787
	Testing Data set (N = 550)
Linear Regression	0.760	0.922	0.773	0.420	−0.104	−10.400%	0.664	0.878	0.771
LR-SVM	0.781	0.930	0.789	0.403	−0.109	−10.900%	0.635	0.890	0.792
LR-RSS	0.934	0.982	0.910	0.179	0.179	0.100%	0.349	0.968	0.937
LR-REPTree	0.993	0.998	0.987	0.041	−0.010	−1.000%	0.109	0.997	0.994
LR-M5P	0.810	0.937	0.753	0.412	−0.132	−13.200%	0.591	0.917	0.841