"Sofia Hggberg" <mpmpmp@xmail.com> wrote in message
news:koapfm$cvq$1@newscl01ah.mathworks.com...
> Hi all,
>
> Could someone please explain me how a person does a dimension reduction
> (for example with PCA) of two multivariate Gaussian distribution and yet
> show that the distance between them is still preserved (they are
> separated).
This seems like less of a MATLAB question and more of a general statistics
question.
If you know of an algorithm to do this and need help implementing it please
post the algorithm (or at least its name or the skeleton of the algorithm)
and someone may be able to help you here.
If you're looking for an algorithm to do this, you probably want to ask on a
statistics-oriented newsgroup like sci.stat.math (accessible via Google
Groups, http://groups.google.com, if your news server doesn't carry it.)
Once you've determined the algorithm you want to use and have started
implementing it, see my previous statement.
--
Steve Lord
slord@mathworks.com
To contact Technical Support use the Contact Us link on
http://www.mathworks.com