Skip to main content

Table 9 Evolution of the standard deviation of residuals for various GRNN models predicting F a and F v in the NP–RF case

From: Deriving amplification factors from simple site parameters using generalized regression neural networks: implications for relevant site proxies

Normalized profiles F a F v
Number of parameters Explanatory parameters Standard deviation of residuals Variance reduction Standard deviation of residuals Variance reduction
All (5) Depth + f 0 + V sm + C v  + V s30 0.0055 99.74% 0.0011 99.98%
3 (best triplet) f 0 + C v  + V s30 0.012 98.7% 0.0034 99.8%
2 (best pair for Fa) f 0 + C v 0.0318 91.43% 0.0172 95.7%
2 (convenient pair) f 0 + V s30 0.0592 70.28% 0.0153 96.6%
1 (best for Fa) C v 0.0783 48.02% 0.0733 21.8%
1 (best for Fv) f 0 0.0944 24.44% 0.048 66.4%
1 (usual) V s30 0.0926 27.3% 0.0709 26.8%
Initial σ 0.1086 0.0829