🚀 go-pugleaf

	In the near future I'm going to have to do a lot of regression  
analysis of experimental data that colleagues and I have acquired here at  
the Oak Ridge National Laboratory.  I'm convinced that I can derive any  
linear or nonlinear least-squares-fit equation required to represent an  
individual data set using a Mathematica package that another colleague has  
given me.  The problem is that I need to determine quickly how well a  
particular equation fits a given set of data, and whether the equation is  
sufficiently well behaved mathematically to "accurately" predict the value  
of the dependent variable in a region (or regions) where experimental data  
have not been obtained.  Fortunately, most of the equations I need to  
derive will have only two "primary" independent variables; consequently,  
both the data and a surface representing the fitted equation can be  
plotted on a 3-D diagram.
	In view of the requirements outlined above, this is what I'd like  
to be able to do.  I would like to buy a piece of software that would  
allow me to view "X-Y-Z" data (accompanied by X, Y and Z axes) in 3-D on  
my monitor, and to rotate the data (along with the three axes) in real  
time.  Furthermore, I would like to be able to zoom into any region of the  
plot in order to get a close-up view of the data and/or surface.  I need  
to be able to use different colors for different data sets, so that I can  
tell whether the equation is fitting one data set better than another.  I  
also need to be able to scale the plot in a way that permits me to assess  
how well the surface extrapolates.
	I'm aware of the distinct possibility that SAS and/or S+ will do  
what I've described above.  However, both of these packages are more than  
I need (obviously), and they're both pretty expensive (especially SAS).   
I'm also aware that I can do much of what I'd like to do using  
Mathematica, but from the knowledge I currently possess, it is my judgment  
that it isn't going to be either quick or convenient to use this program  
either.  I have no knowledge of what X programs might be out there to do  
what I need to have done.  If there is such a program (and it's not too  
expensive--say, no more than ~$600), I would be willing to buy it and run  
it using CoXist.
	Obviously, I badly need advice from some kind soul in netland.  An  
e-mail response would be great, or call me at (615) 574-7041, or e-mail me  
your phone number and I'll be happy to call you.  Thanks in advance.

Jim