Problem using FSOLVE 'OptimalityTolerance' is not an option

Hi

I'm trying to do Nonlinear regression with Levenberg Marquardt, but I have this problem.

Error using optimoptions (line 105)
'OptimalityTolerance' is not an option for FSOLVE.
A list of options can be found on the FSOLVE documentation page.

Error in Transfer_function_corrected (line 116)
        options =
        optimoptions(@fsolve,'TolFun',1e-54,'Algorithm','levenberg-marquardt','MaxIter',1e16,'Display','off','TolFun',1e-12,'OptimalityTolerance',1e-12);

My code..

[l,a]=size(H);
for y=1:a
    for z=1:l
        myfun = @(x) [((H(z,y)-(D(y)*exp(-1i*x(1)*(r(y+1)-r(1))))))^2];
        options = optimoptions(@fsolve,'TolFun',1e-54,'Algorithm','levenberg-marquardt','MaxIter',1e16,'Display','off','TolFun',1e-12,'OptimalityTolerance',1e-12);
        [K(z,y),fval(z,y)] = fsolve(myfun, [0],options);
        c=c+1;
        waitbar(c/n)
    end
end

Thank you so much.

Answer 1

[Alan Weiss]

I am surprised that you got that error, but here is what the fsolve documentation has to say about that option:

Internally, the 'levenberg-marquardt' algorithm uses an optimality tolerance (stopping criterion) of 1e-4 times FunctionTolerance and does not use OptimalityTolerance.

I would also like to point out that many of your chosen options and tolerances do not make sense. To quote from that link: Generally set tolerances such as OptimalityTolerance and StepTolerance to be well above eps, and usually above 1e-14. Setting small tolerances does not always result in accurate results. Instead, a solver can fail to recognize when it has converged, and can continue futile iterations. A tolerance value smaller than eps effectively disables that stopping condition.

Alan Weiss

MATLAB mathematical toolbox documentation