Least squares regression - removing insignificant terms quickly?

[#1]

Hi all-

I am doing least squares fits, up to cubic with 15 X's. Only the 1st several ever turn out "significant", so those at the "bottom" with small p values could be removed...one at a time...as long as R^2 doesn't go down. This would be a slow iterative process, but would result in a much shorter equation. I know of one other tool that will do this.

Can JMP?

Thanks,
Dave

[#2]

> I still wonder if the "backwards"
> elimination of insignificant terms actually gives a
> more phsysics based model. If so, I'd prefer that.

Backwards elimination (and other statistical methods) do not result in a "phsysics [sic] based model". These statistical modeling techniques give you an empirical model, and don't even attempt to produce a"phsysics [sic] based model". These statistical modeling techniques give you a model based upon the data, based upon the algorithm you use, and the algorithm parameters that you specify.

Do not think that the model you come up with via statistical methods represents true underlying physical mechanisms — the model may, or may not, represent the underlying physical mechanisms.