Timeline for "Normalizing" variables for SVD / PCA
Current License: CC BY-SA 3.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 20, 2015 at 0:08 | comment | added | amoeba | -1 Downvoted, because this answer is only about centering, whereas the question is about scaling. Also, the claim that the first eigenvector is necessarily the vector of means is wrong, as noted by @whuber. | |
| Jul 9, 2014 at 20:46 | comment | added | whuber♦ | When you don't center the samples (subtract the means from the columns), the first eigenvector usually is not the vector of means. | |
| Jun 22, 2011 at 7:43 | comment | added | Szabolcs | I realize there might not be a rotation-invariant way to do it, but I'd love to at least read some discussion of these issues ... any pointers welcome. Note: I have no training in applied stat (only maths, such as linalg, prob theory), so I'm learning this stuff as I'm going. | |
| Jun 22, 2011 at 7:41 | comment | added | Szabolcs | One might simply divide each variable by its standard deviation, but I was wondering if there are other things people do. For example, we can think of this dataset as a point cloud in $N$-dimensional space. Is there a way to do it in a way that does not depend on the rotation in this $N$-d space? If we divide by standard deviations, it will matter along which axes those standard deviations are taken (i.e. it's not rotation invariant). If we do it along the principal axes, then I think the variables will appear uncorrelated. | |
| Jun 22, 2011 at 7:37 | comment | added | Szabolcs | I should have mentioned in the question that the mean has already been subtracted. I'll edit it accordingly. | |
| Jun 22, 2011 at 4:44 | history | answered | petrichor | CC BY-SA 3.0 |