Gents, thx again for reply.
Cummings, the 2nd order polynomial, which has more than 7 coefficients, works, thus, computational power may not be the issue when talking about 16 coefficeints, for a 3rd order, 2 independent varaibles polynomial regression.
Kiwi and Clyde, thx for info, but what I...
Gents, thx for your suggestions.
Cummings, unfortunatelly, for what I need, I cannot use a commercial package, thus, I'm stuck with Excel even for the non-linear regressions.
Thus, regarding power, [y=a*(X1)^b1*(X2)^b2], I used
LINEST(LN(A1:An),LN(B1:Cn),TRUE,TRUE), where y is in column A, X1...
Cummings,
Thank you very much for your suggestion, it works. I also figured out how to get all the coefficents of all the other terms, following the pattern in your formula.
Now, I have to take care of the power regression, of the type y = a*(x1^b)*(x2^c). Any suggestions for this one?
Dan
Cummings,
Thanks for help, it started to work. I tried the LINEST() function as you described it and got the coefficients for x1, x1^2, x1^3, x2, x2^2, x2^3, and the free one, but without any terms for products between the 2 independent variables (e.g.: x1*x2, x1^2*x2, x1^3*x2, etc.
I checkd...
Cummings54, thx for your suggestion. However, it applies to single variable polynomial regression, not multivariables. What I need is multivariable polynomial regression (and multivariable power regression), of the type y = f(x1,x2...xn). To simplify, let's say omly 2 indepemdent variables, 3rd...
Denial, thank you, I'll try it. I used LINEST() for linear multivariable, and single variable linear, polynomial, etc., but it hasn't crossed my mind to build extra columns with x1^2, x1^3, and respectively x2^2, x2^3 to use it for multivariable polynomial regression.
Dan
Does anyone have any suggestions how to do a multivariable polynomial (2nd and 3rd order) and power regressions in Excel and calculate the coefficients and some statistical characteristics of the fitted equation? Let's assume for simplicity only 2 independent variables.
The second question is...
