Hello all,

I remember a discussion taking place here about a year ago
regarding signal processing issues and the use of cubic and
quintic splines.
However, what wasn't mentioned (I don't
think) was the possibility of applying splines to
multivariate problems ie. fitting polynomial surfaces (and
hyper-surfaces) rather than just polynomial lines.
Can this be done, and if so how??

I have already multiple linear regression techniques to
perform something similar, but this applies just one
polynomial surface to the complete set of data rather than
a set of surfaces.

Any pointers will be highly appreciated!
Many thanks,

Nick Barnett.


__________________________________________________ ____________________
Nick Barnett.
Centre for Rehabilitation and Engineering Studies (CREST),
University of Newcastle-Upon-Tyne,
Newcastle-Upon-Tyne, e-mail ......: N.D.Barnett@ncl.ac.uk
England. Telephone ...: +44 (0)191 222 5434