0
$\begingroup$

In my textbook it says the data is assumed to be zero-centered. I am not sure what this means. I read it stand for substracting the mean from the individual values, however I would like to know do you divide the values by the standard deviation as well.

For example if one dimension had values 8 10 12 and another 80 100 120, the latter would have much more variation and consequently PCA would assign it a greater weight. My question is whether we divide the data by standard deviation as well.

$\endgroup$
3
$\begingroup$

You're right, each attribute should always be normalised before PCA occurs. Check out this answer: Why do we need to normalize data before principal component analysis (PCA)?

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.