Table 4.

Parameters and prediction differences of clavulanate models.

Parameter	Clavulanate Model 1		Clavulanate Model 2		Clavulanate Model 3		Clavulanate Model 4
	Training set	Test set	Training set	Test set	Training set	Test set	Training set	Test set
Number of samples	53	16	46	19	30	12	32	10
Content range (%, mg/mg)	1.13–17.62	1.40–15.79	1.13–20.12	5.18–17.61	1.10–19.60	1.16–14.63	1.03–3.14	1.29–3.01
Wavenumber range (cm−1)	7702.5–7124.0,6549.3–5970.7		10001.3–8848.1, 6549.3–5396.0		6549.3–5970.7,4825.2–4246.6		10001.3–8848.1,6549.3–5396.0
Pre-processing method	1st derivative + vector normalization		1st derivative + vector normalization		1st derivative + vector normalization		1st derivative + vector normalization
Rank	5		6		6		6
R2 (%)	99.63		99.48		99.63		99.78
RMSEP (%)	0.271		0.386		0.226		0.0317
Samples whose prediction difference is >5%	1.8% (7/377)		1.8% (7/377)		0% (0/78)		0% (0/96)
Samples whose prediction difference is <1%	69.0% (260/377)		72.1% (272/377)		60.0% (47/78)		93.7% (90/96)
Average prediction difference	0.8%		0.9%		0.8%		0.2%