The coefficient of determination also known as R^2 tells how good a fit is. If R^2=1 the fit is perfect an if R^2=0 it's useless. But Maple don't have a native function to calculate R^2. I seached and found this:
But it only describe how to calculate R^2 on a linear fit. I can understand from this:
"Users of linear regression models are accustomed to expressing the quality of fit of a model in terms of the coefficient of determination, also known as R2. In nonlinear regression, such a measure is unfortunately, not readily defined. One of the problems with the R2 definition is that it requires the presence of an intercept, which most nonlinear models do not have. A measure, relatively closely corresponding to R2 in the nonlinear case is Pseudo-R2 = 1 - SS(Residual)/SS(TotalCorrected)." source:http://www.ats.ucla.edu/stat/sas/library/sasnlin_os.htm#Calculating a R2-Type Measure
That there is no real R^2 for non linear fits but there exist a Pseudo-R^2.
My question is: how do i calculate (Pseudo-)R^2 in a non-linear fit?
Also in excel there is an option to calculate R^2 on a non-linear fit: