CODEWITHSUNDEEP
matlab
user30609
user30609

Reputation: 264

error in objective function of fmincon and solver stopped prematurely, MATLAB for loop

I am completing maximum likelihood estimation inside a for loop.

When I don't converge, I get this error message,

error in objective function of fmincon and solver stopped prematurely

I have solved this with this answer: https://uk.mathworks.com/matlabcentral/answers/131975-error-in-objective-function-of-fmincon-and-solver-stopped-prematurely

However,

on certain occasions, I still can't converge.

Is it possible to count (automatically) how many times I don't converge?

Upvotes: 2

Views: 518

Answers (1)

obchardon
obchardon

Reputation: 10792

You can use the exitflag output of the fmincon function:

[x,~,exitflag] = fmincon(fun,x0,A,b);

exitflag indicate the reason why fmincon stopped.

According to the documentation exitflag can take the following value:

All Algorithms:

 1 %First-order optimality measure was less than options.OptimalityTolerance, and maximum constraint violation was less than options.ConstraintTolerance.
 0 %Number of iterations exceeded options.MaxIterations or number of function evaluations exceeded options.MaxFunctionEvaluations.
-1 %Stopped by an output function or plot function.
-2 %No feasible point was found.

All algorithms except active-set:

 2 %Change in x was less than options.StepTolerance and maximum constraint violation was less than options.ConstraintTolerance.

trust-region-reflective algorithm only:

 3 %Change in the objective function value was less than options.FunctionTolerance and maximum constraint violation was less than options.ConstraintTolerance.

active-set algorithm only:

 4 %Magnitude of the search direction was less than 2*options.StepTolerance and maximum constraint violation was less than options.ConstraintTolerance.
 5 %Magnitude of directional derivative in search direction was less than 2*options.OptimalityTolerance and maximum constraint violation was less than options.ConstraintTolerance.

interior-point, sqp-legacy, and sqp algorithms:

-3 %Objective function at current iteration went below options.ObjectiveLimit and maximum constraint violation was less than options.ConstraintTolerance.

If fmincon sucessfully determine a minimum then exitflag == 1. So you can simply count how many time exitflag ~= 1

error_counter = 0;
for ii = 1:n_iteration
    [x,~,exitflag] = fmincon(fun,x0,A,b);
    if exitflag ~= 1
        error_counter = error_counter + 1
    end
end

And if you only want to check that the number of iterations has not been exceeded then you can replace exitflag ~= 1 with exitflag == 0.

Upvotes: 4

Related Questions