Improving Accuracy of River Flow Forecasting Using LSSVR with Gravitational Search Algorithm

Table 6

Comparison of the LHLGSA, LMARS, LM5RT, and LCMLR methods: Shyok catchment.

Indexes	Cross validation data sets	Input combinations	Methods
			LHLGSA	LMARS	LM5RT	LCMLR
	DS1	IC1	165.03	224.25	240.12	395.55
		IC2	142.12	160.35	163.1	195.03
		IC3	153.55	157.09	171.48	167.49
	DS2	IC1	154.19	210.33	224.47	470.32
		IC2	133.53	172.82	183.1	201.85
		IC3	134.43	155.31	174.73	171.76
	DS3	IC1	146.79	186.3	207.49	370.18
		IC2	109.41	120.66	141.03	162.35
		IC3	117.29	122.36	124.84	138.19
	DS4	IC1	207.08	243.15	259.96	512.25
		IC2	161.58	171.43	185.77	208.59
		IC3	163.91	168.28	209.68	191.93
	DS1	IC1	77.85	101.03	131.89	212.74
		IC2	65.08	73.75	78.02	85.02
		IC3	66.44	66.09	102.13	66.82
	DS2	IC1	83.19	99.06	111.59	230.22
		IC2	78.05	82.59	94.03	88.69
		IC3	78.8	80.38	92.82	86.81
	DS3	IC1	75.99	90.21	101.15	190.72
		IC2	59.01	64.28	72.03	76.79
		IC3	65.47	69.9	71.92	74.58
	DS4	IC1	102.66	124.59	138.56	240.1
		IC2	76.67	82.41	84.61	93.64
		IC3	76.89	80.77	105.49	84.59
	DS1	IC1	0.845	0.721	0.738	0.677
		IC2	0.896	0.884	0.879	0.869
		IC3	0.89	0.888	0.842	0.873
	DS2	IC1	0.896	0.819	0.818	0.694
		IC2	0.91	0.898	0.861	0.856
		IC3	0.904	0.901	0.884	0.896
	DS3	IC1	0.921	0.782	0.899	0.721
		IC2	0.959	0.947	0.923	0.911
		IC3	0.953	0.938	0.930	0.920
	DS4	IC1	0.823	0.752	0.736	0.625
		IC2	0.89	0.887	0.88	0.832
		IC3	0.873	0.895	0.782	0.855