I have a 3-D dataframe representing family of curves: yi=f(xi) which are also depending on a third variable lets say zi (yi=g(xi,zi)) as shown in the graphic below:
The yi data are measurements and each curve is valid on its own.
Now I would like to find out a proper way to fit a plane or surface to this data in order to avoid 2-D bilinear data interpolation.
In general I am trying to do a similar approach as in the link below:
With the difference that my x and y data do not have the same dimensions. What I am looking for is more a solution for this kind of problem: