Is it possible that the standard deviation of one or some of the Principal Components obtained from features are more than any of the features.
For eg. If my standard deviation for features: feat1, feat2, feat3, feat4, feat5, feat6 are 0.019, 0.027, 0.026, 0.025,0.026,0.030,0.019. I have obtained the Standard Deviations for the Principal Components as: PC1, PC2, PC3, PC4, PC5, PC6 as 0.05, 0.020,0.018, 0.016,0.014,0.012
As you can see PC1 has higher standard deviation than the rest. Is this possible ?
Is it possible that the standard deviation of one or some of the Principal Components obtained from features are more than any of the features.
Yes, and that is the sole purpose of PCA. We want to find a set of orthogonal axes along which the variance (and therefore the standard deviation) of the data set is maximized.
See the explanation here for more