PCA on binary dataset

feature_engineering
dimensionreduction

#1

Hi guys,

I am working on a dataset with 1500 binary variable. I am wondering if it is a good idea to apply PCA treating it as continuous data or use MCA or Factor Analysis.


#2

You can use MCA and also categorical principal component analysis.
https://www.researchgate.net/file.PostFileLoader.html?id=5798ef16f7b67e72863fad65&assetKey=AS%3A388514446364673%401469640470459


#3

okay, thanks. I was trying to understand why doesnt PCA work in case of binary variables and can we make some changes before creating the eignevectors to get a better matrix which we can apply if we have binary dataset.

I will try MCA and update here .