A New Method for Spectral Wavelength Selection Based on Multiple Linear Regression Combined with Ant Colony Optimization and Genetic Algorithm

Table 2

The results of different methods on the grain protein dataset.

Method	RMSECV^a	RMSEC^b	nVAR^a	nVAR^b
Full	0.6288	—	117	—
SPA	0.3943	—	27	—
CARS	0.3231  0.0231	0.2870–0.3900	30.93  6.86	22–40
ACO	0.2889  0.0060	0.2805–0.3088	31.66  4.28	21–43
GA	0.3159  0.0102	0.2791–0.3332	54.2  45.37	43–66
DE	0.2985  0.0035	0.2876–0.3047	49.43  4.26	42–60
MLR-ACO	0.2689  0.0049	0.2552–0.2759	40	—
MLR-ACO-GA	0.2674  0.0043	0.2552–0.2759	40.46  1.33	37–43

^aStatistical results of mean ± standard deviation of 30 replicate simulations of different methods. ^bStatistical results of the variation range of 30 repeated simulations with different methods.