Skip lsqnonlin run if error

Hi guys. I am using lsqnonlin for some fitting, and I am running a loop where I change the initial parameter values to determine the best ones based on the residuals; this is because the fitting is not as robust as I expected and it gives different results for different initial parameters (I would love to hear a suggestion for this as well...!). So, for some combinations of parameters it fails and gives this error:

Error using levenbergMarquardt (line 16)
Objective function is returning undefined values at initial point. lsqnonlin cannot continue.

How can I make the process recognize the error, skip this run and proceed with finishing the loop?

Thanks

Answer 1

[Alan Weiss]

If you have Global Optimization Toolbox, then use MultiStart to run lsqnonlin several times. MultiStart automatically continues if it encounters NaN or Inf values.

Otherwise, it is not too hard to program up this kind of functionality for yourself. If you have finite bounds on all parameters, you can repeatedly use

x0 = lb + rand(size(lb)).*(ub - lb)

as initial points. You can use a try/catch statement to continue after encountering an error.

Good luck,

Alan Weiss

MATLAB mathematical toolbox documentation

[Xen]

The try/catch method did the trick! I will try the other suggestions soon. Thanks Alan.