MAPLE Users,

Suppose I have a set of points in some N-dimensional space.  I would like to obtain a simple polynomial that

best fits the data.  I do not know in advance the form of the function, but a simple function that does not overfit

the data would probably be OK.  By "simple" I mean the smallest degree with or without cross-terms that gives a

decent fit.  My data set will typically be an external comma or tab separated text file.