# Dimensionality Reduction. How to explain dynamics of feature subset based on all features data?

I have features: f1..f1000. I want to explain dynamics of particular features subset: f1-f5 based on all features data (based on f1..f1000).

Preferably I think of something like "partial PCA", where some Principal Components created on all features will explain dynamics only of 5 of them (not all of them).

If you know any kind of such Dimensionality Reduction technique I would be very glad for your help. Thanks!

• I'm not sure what you are asking: What do you mean by explaining dynamics of a featureset? – S van Balen Aug 15 '19 at 11:02
• Ok, that's a fancy way of saying - explain variance of the subset of features by all features – Ziemo Aug 15 '19 at 19:08