Convert two categoricals variables to numeric one - PCA FactoMineR

Asked Jan 22 '20 at 16:59

Active Jan 22 '20 at 21:10

Viewed 15 times

I am currently working on the CPA with R and I have a problem. My data are all numerical, except for two columns that represent a "category" (the sampling site and the land use type) At first I foolishly replaced these data with numerical identifiers (A:1, B:2, C:3 ...), but this method implies that A < B < C. I don't know if this is really a problem for PCAs, but A is definitely not < B nor B < C.

Is there a way to convert a text data into a numerical data without biasing the results to make a PCA ?

Thank you.

edited Jan 22 '20 at 21:10

asked Jan 22 '20 at 16:59

user3424481

1

https://stats.stackexchange.com/questions/5774/can-principal-component-analysis-be-applied-to-datasets-containing-a-mix-of-cont – Janosch Jan 22 '20 at 22:04
One hint: CATPCA – ttnphns Jan 22 '20 at 23:02

Convert two categoricals variables to numeric one - PCA FactoMineR

0 Answers0