PCA scores in my own implementation have different sign from the ones computed in R

Question

I am trying to implement PCA (NIPALS method) using Excel. The problem is that my results are not exactly matching with the results of R. There is only a difference in sign: if my scores are positive, then scores using R are negative.

I don't know what is going on. What to do?

score 1 · Answer 1 · answered Sep 04 '13 at 12:52

1

1) I wouldn't use Excel for data analysis

2) The signs of PCA are arbitrary, so there is no real problem here.

answered Sep 04 '13 at 12:52

Peter Flom

119,535
36
175
383

Oh, Is it so. Ill see to it. Thank you so much Peter. – Sh_sharma Sep 04 '13 at 12:57

PCA scores in my own implementation have different sign from the ones computed in R

1 Answers1