finding Optimum point among many points

I have 8 set of measurement data from a current sensor with differing values of current. I fitted them to a equation and found 8 solution (using fmincon) to corresponding data set. Now I want to combine these 8 solutions and merge them into one solution(for all current values). I can understand that the fit level for the resulting solution may be less but how to find the final best solution that can fit all my 8 data set?

Is there any other way to find an optimum point for the 8 data set directly? i.e fitting an equation to 8 dataset directly...

[Alan Weiss]

I am not sure that I understand you. Are you looking for the point that makes the smallest total error for the 8 data sets? I mean, the total error is the sum of all the squared residuals.

If you mean something else, please state exactly what you are looking for, maybe in an equation.

Alan Weiss

MATLAB mathematical toolbox documentation

[praveen]

Yes what you said is correct i want very less sum of error for all data sets combined.

Answer 1

[Alan Weiss]

If I understand you correctly, you are looking for a least-squares solution for a nonlinear model. Most likely, lsqnonlin or lsqcurvefit would solve your problem. See, for example, Nonlinear Curve Fitting with lsqcurvefit.

Alan Weiss

MATLAB mathematical toolbox documentation

[praveen]

Thanks dude