1

I just asked a question about PCA and scipy's implementation of PCA compared to mine. You can find it here:

https://stackoverflow.com/questions/38799205/why-are-my-manual-pca-reconstructions-not-matching-pythons-sklearns-reconstruc

It seems the question could have been very suited for this site. How do I know if this question specifically was better for stats.overflow, stackoverflow or computation science overflow.

2

It depends on the question you are trying to ask. If your question is an algorithms question, Stats or SciComp would be good fits. If you want a statistics perspective on PCA, ask Stats. If you want a more general computational science perspective on PCA, ask SciComp. Sometimes, SciComp users are also willing to help with debugging; I am not sure about Stats, since I don't spend much time there.

StackOverflow, as far as I can tell, is mostly about programming. There are people who answer math questions over there, but the site isn't devoted to math algorithms; it's devoted to programming and software development. So if you want a programming or software development perspective on PCA, ask StackOverflow.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .