Table 1 Amounts of variation of XORTH,XOSC, and Y in the OPLSR method for the simulation study are shown. The first OSC component and the first PLS component accounted for more than 96% of the total X variation and 99% of the total XOSC, respectively. The first PLS component accounted for more than 99% of the Y variation in the simulation study
From: Feature selection using distributions of orthogonal PLS regression vectors in spectral data
| Â | Â | Â | PLS | Â | Â | PLS | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Â | OSC | Â | 1 | 2 | 3 | Â | OSC | Â | 1 | 2 | 3 |
XORTH | 1 | 0.967 | Â | Â | Â | Â | 2 | 0.967 | Â | Â | Â |
XOSC | Â | 0.033 | 0.991 | 0.002 | 0.002 | Â | Â | 0.033 | 0.994 | 0.002 | 0.001 |
Y | Â | 1.00 | 1.00 | 0.000 | 0.000 | Â | Â | 1.00 | 1.00 | 0.000 | 0.000 |