I am hoping you guys can point me in the right direction for what I should go read more on.
Goal: I have a sample data set of 3D matrices. I want to determine if a new data point (3D matrix) is within the bounds of the original data set.
Is there some sort of unique property or some way to characterize these matrices?
I thought maybe I could approximate a characterization with the first 4 moments of the distribution, but I am wondering if there is something smarter and more exact out there that I just dont know about.
I am not a mathematician but have studied similar sounding things in grad school economics and stats classes.
It is not clear to me what a data point in terms of a 3D matrix is. Could you clarify or given an example?