Principal Components Analysis With JMP 

From the presenter in this video, Kemal Oflus: JMP provides 3 options to calculate principal components: - Using the Correlations, - Using the Covariances, and - Unscaled The calculation for the principal components depends on which matrix you select for extraction of principal components: - For the on Correlations option, the i-th principal component is a linear combination of the centered and scaled observations using the entries of the i-th eigenvector as coefficients. - For the on Covariances options, the i-th principal component is a linear combination of the centered observations using the entries of the i-th eigenvector as coefficients. - For the on Unscaled option, the i-th principal component is a linear combination of the raw observations using the entries of the i-th eigenvector as coefficients In other words, the coefficients that are in the principal components formula are the coefficients in the eigenvector matrix (adjusted by the selection of the options above). If you run the principal components analysis with the “Unscaled” option, the coefficients for the eigenvectors will be the same as the values in the principal components formula. Otherwise, the data is centered and scaled; that’s why the values in the eigenvector matrix are different than the principal components formula.

Excellent, Thank you

Hi, I don't really get how did you get those values in the PC1 formula? Could you tell me where I could find these values ?

Thanks for the video, it's very useful. I have a question. In the formula, why is more than 1 the addition of the percentages of the values conforming the Principal Component??

Why does the coefficients of the formula of Prin 1 not match with your eigenvectors of Prin1?

cant understand what he mumbles about. thumbs down.