How can I make the estimation more stable?

function [a, fval, exitflag, OUTPUT] = Estimator(rr, lambda)

Estimation of parameters teta=[t1min t2min tp1 tp2] by jointly maximizing the log-likehood function with respect to teta Algorithm used: ______________

t = [ rand(1) rand(1) rand(1) rand(1) rand(1)]; %Initial Guess
options = psoptimset('CompleteSearch', 'on', 'MaxIter', Inf, 'MaxFunEvals', 1000*5);%1e1000000*5
% options = gaoptimset('Generations', 1e20);
% lambda = data(6);      % As lambda we are using the set parameter

Lb = zeros(1,5);
Ub = ones(1,5);
Aineq = [1 -1 0 0 0; 0 0 0 0 1];
bineq = [-.1; 1];

JPF = @(t) AV_function_eps(t, rr, lambda);

[a, fval, exitflag, OUTPUT] = patternsearch(JPF, t, Aineq, bineq, [], [], Lb, Ub, options);

I have noticed that if I repeat the estimation several times I obtain different results.

Answer 1

[Alan Weiss]

I am sorry, I do not understand your question. What exactly are you asking? What are you assuming?

Also, if you ask again, please format your question using the tools to format your code, the {}Code button. Formatting can make code much easier to read.

Alan Weiss

MATLAB mathematical toolbox documentation