1

Could someone please explain me this reviewer comment:

asymmetric distribution could affect Principal Component Analysis results, symmetry of distribution should be tested. Authors should also indicate if outliers were observed and consequently excluded because they could affect factors

My question: what does it mean asymmetry distribution could affect PCA? And also outliers, could they affect factors?

kjetil b halvorsen
  • 63,378
  • 26
  • 142
  • 467
user22046
  • 11
  • 1
  • 2
    Yes, asymmetry and outliers can affect the PCA decomposition. For an example, have a look [here](http://stats.stackexchange.com/a/33602/603) – user603 Mar 15 '13 at 14:04
  • 1
    See also this [question](http://stats.stackexchange.com/q/38709/3277). – ttnphns Mar 15 '13 at 17:14

0 Answers0