A Conjunction Method of Wavelet Transform-Particle Swarm Optimization-Support Vector Machine for Streamflow Forecasting

Table 3

Forecasting performance indexes of SVM and WSVM models in Tangnaihai station.

Model	Training				Testing
	𝑅	RMSE (m^3/s)	MAE (m^3/s)	MARE (%)	𝑅	RMSE (m^3/s)	MAE (m^3/s)	MARE (%)
(i) SVM	0.853	287.315	196.524	14.33	0.768	317.035	247.480	17.81
(ii) WSVM1	0.548	306.554	217.883	18.24	0.613	331.420	256.368	23.49
(iii) WSVM2	0.808	201.784	141.730	7.46	0.806	243.268	173.20	11.52