International Journal of Concrete Structures and Materials

Table 4 SPCA models’ R², \(R_{a}^{2}\), \(\lambda\), RMSE, and MAE values.

From: Prediction of Stress Increase at Ultimate in Unbonded Tendons Using Sparse Principal Component Analysis

Variables	Internal data					External data
Variables	\(R^{2}\)	\(R_{a}^{2}\)	\(\lambda\)	RMSE	MAE	\(R^{2}\)	\(R_{a}^{2}\)	\(\lambda\)	RMSE	MAE
All variables	0.46	0.46	0.99	107.93	85.83	0.70	0.69	0.99	152.79	110.93
Cont. and Cate.	0.44	0.43	1.02	109.72	88.58	0.70	0.69	0.99	152.79	110.93
Self-Selected	0.31	0.30	0.95	121.96	101.16	0.54	0.52	1.05	189.29	151.38
Corr. Cutoff	0.54	0.53	1.03	99.53	78.04	0.68	0.68	1.00	156.88	113.58

Back to article page