I am trying to curve fit (line regression) a curve that is a function of three variables. I have successfully plotted other curves with bi-quadratic methods using LINEST, but I cannot figure out how to do this with a function that contains three variables.

For two variables I have 6 columns; Outdoor Air Temp, Wetbulb Temp, and Superheat given in the form of:
OAT, OAT^2, WB, WB^2, OD*WB, SH.
I use LINEST(SH1:SH13,OAT:1:OD*WB:13,,TRUE) and I come out with a VERY accurate representation of the curve.

How can I use this function for three-variables..is it possible?