Fig. 9From: A dimension reduction technique applied to regression on high dimension, low sample size neurophysiological data setsCV errors for the proposed regression method (RoLDSIS) and the methods of regularized linear regression. The results for the physical (\(\Phi\)) and the psychophysical (\(\Psi\)) attributes are shown in the top and the bottom panels, respectively. The mean squared errors for the test set of the CV are shown with dots. Confidence intervals at 95% are represented by vertical barsBack to article page