I am using PCA to do the data inspection. First 3 PCs explain nearly 82% of the total variance. Suppose the number of features is $n$. And I found 4 of the features have similar PCA feature importances for first 3 PCs. Say in the attached figure: enter image description here evap_c_5, evap_c_7, w_s_5, w_s_7 have similar feature importance for pc_1, pc_2 and pc_3. So can I safely conclude that those 4 variables are similar in the original space? if so, could you provide a theoretical derivation, or a literature for reference? Thanks!



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.