0
$\begingroup$

I am dealing with a database where frequencies of behaviors are recorded, thus being continuous data but with many zeros. Aiming to reduce the dimensions of the seven variables, I have carried out a principal component analysis. I was wondering if this methodology is appropriate for this kind of data set, or whether it is preferable to use other methods such as NDMS or PCoA.

$\endgroup$
2
  • $\begingroup$ Could you explain your intended meanings of "appropriate" and "preferable"? After all, PCA will always succeed in reducing the dimensions. But that's almost never the objective of an analysis: dimension reduction is a preliminary step towards something else. What do you want to do? $\endgroup$ – whuber Apr 12 at 14:55
  • 1
    $\begingroup$ We aimed to apply the PCA to reduce the number of behavioral variables recorded for our species model. Secondly, each PC would be considered as a response variable of a GLMM. The doubt lies in whether applying a PCA for a database with many zeros may bias the results obtained in the first step. $\endgroup$ – Jfgrass Apr 12 at 15:53

Your Answer

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

Browse other questions tagged or ask your own question.