Chronological Age Estimation of Male Occipital Bone Based on FTIR and Raman Microspectroscopy

Table 3

Results of PLS and GA-PLS regression model of chronological age estimation. The optimized models are shown in bold.

Instrument	Method	VNs	LVs	Cross-validation		Prediction
				R^2CV	RMSECV	R^2Pred	RMSEP
Eight outcomes	PLS	8	1	0.474	16.1257	0.383	20.4607
FTIR	PLS	259	8	0.710	13.1824	0.731	12.1969
	GA-PLS	32	9	0.814	10.405	0.828	9.2654
Raman	PLS	1557	4	0.657	14.0765	0.627	13.3943
	GA-PLS	196	10	0.770	11.6277	0.738	11.0321

VNs: variable numbers; LVs: latent variables; R²CV: determination coefficient of cross-validation; RMSECV: root mean square error of cross-validation; R²Pred: determination coefficient of prediction; RMSEP: root mean square error of prediction.