Can I mix different data transformations for different variables in the same dataset? Target analysis: Principal Component Analysis.

  • $\begingroup$ can you give an example of what you mean? $\endgroup$ – gunes Mar 22 at 8:54
  • $\begingroup$ @gunes for example, can I use log transformation on some variables and square root transformation on other variables in the same dataset? $\endgroup$ – f.catus Mar 23 at 18:02
  • $\begingroup$ Please search our site for (very) closely related discussions. $\endgroup$ – whuber Oct 12 at 13:46

There is nothing in the mathematics of Principal Component Analysis that prohibits the use of different transformations on diferent of the input variables. So you can do it. If it makes sense depends on details of the actual application you did not tell us.

| cite | improve this answer | |

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.