I have some 2-dimensional data to which I would like to fit a curve. The data is not functional, i.e. there can be multiple values of y for a given x. The curve should be continuous, but otherwise I have little idea about what form it should take.
Maple provides a really nice way to plot one such curve: Statistics[ScatterPlot] with the lowess option. But I do not want to just plot the curve. I would also like to do computations with the curve. For example, I would like to compute the sum of squared residuals.
What is the best way to get the curve (as a procedure, say)?
Also, what is the reason that lowess is an option of a procedure in Statistics[Visualization]? Would it not be more in keeping with Maple paradigms for lowess to be a separate procedure in Statistics[DataSmoothing]? That would have solved my problem. (The Help for Statistics[DataSmoothing] even illustrates ScatterPlot with lowess!)