I have a very noisy set of data that I'm trying to analyze. The data consists of several sets of X-Y points. The X points are evenly spaced, but I have multiple sets in an attempt to provide more data to the curve fitting algorithm. So I end up with multiple Y values for each X value. When I attempt to use these arrays in
CWStat1.PolyFit x, y, Order, FitData, CoEff, mse
command, I get junk. If I only have one Y value at each X, then this generates a nice fit. The problem is that once in a while I get a really bad number that throws off the curve. I was hoping that the extra data would prevent this. Any ideas?
