On 8/6/2013 7:23 PM, TideMan wrote: > On Wednesday, August 7, 2013 10:07:12 AM UTC+12, Bo wrote: >> I have experiment data (xdata, ydata) >> >> math model established (with some parameters) >> >> what statement should I use to find solve those parameters with my experiment data and find the best fit line of the model? > Well, it depends................. > Is your model linear or nonlinear? > > If it's linear, you can use least squares regression: > help \ > and read the second paragraph. > > If it's nonlinear, then that's a can of worms and you need to give us an idea of what the model looks like. >
The Statistics Toolbox function nlinfit might be relevant. The Optimization Toolbox function lsqcurvefit might be helpful.
But, as TideMan says, details are important, these may or may not be relevant for you.
Alan Weiss MATLAB mathematical toolbox documentation