Peak Ground Acceleration Prediction by Artificial Neural Networks for Northwestern Turkey

Table 6

R, RMSE, and MAE values of each ANN method for models (v), (vi), (vii), (viii), and (ix) in the test period.

Model	Inputs	Outputs	FFBP method			GRNN method			RBF method
			R	RMSE (cm/s/s)	MAE (cm/s/s)	R	RMSE (cm/s/s)	MAE (cm/s/s)	R	RMSE (cm/s/s)	MAE (cm/s/s)
(v)	𝑀𝑤, FD and HD	PGA for E-W	0.856	44.45	18.46	0.829	46.89	18.63	0.487	55.68	29.79
(vi)	𝑀𝑤, FD and HD	PGA for N-S	0.122	33.22	16.43	0.109	32.95	16.62	0.066	47.27	26.73
(vii)	𝑀𝑤, FD and HD	PGA for U-D	0.093	98.38	93.61	0.933	14.22	5.27	0.751	21.87	10.16
(viii)	𝑀𝑤, FD and HD	Max. PGA	0.895	29.55	15.18	0.882	45.29	19.38	0.563	61.52	37.75
(ix)	𝑀𝑤, FD, HD and SC	Max. PGA	0.667	46.27	21.94	0.803	36.56	15.35	0.07	50.48	95.93