Iterative Optimization with fminsearch

I have 2 time series data s and t. I have a function Error code:

function Error = Output (s,t,k) error = s*k(1)*k(2)*k(3)*k(4)-t; Error = sum(error^2); end

I have initial k, k0 = [a b c d e f]. So 6 parameters to vary. I want to create a loop that will keep fminsearch running for a range of values of k. That is, from k0 = [a0 b0 c0 d0 e0 f0] to k0 = [an bn cn dn en fn]. I also want to put different intialval:step:final value for each of a,b,c,d,e,f. So let a be from 1 to 5 with step of 1, b from 0.1 to 0.5 with step of 0.1 etc If the value of fval is less or equal to 1, I want the iteration or loop to end, otherwise the loop should continue to the end. This is the code I wrote so far and the error it gives me: t = a; s = c;

for k0 = [0.2 0.5 1 1.5]:[0.25 0.6 1.1 1.6]

    [k,fval,exitflag,output] = fminsearch(@(k) suki(s, t, k), k0)

    if fval <= 1
        disp('Convergence has worked.')
    elseif fval >= 1
        disp('The convergence is not good enough')
    end

end The error it gives is: Attempted to access k(2); index out of bounds because numel(k)=1.

Can anyone help me please in implementing this?

[Titus Edelhofer]

Hi Suki,

is this the same question as question?

Titus