Re: Mean-Variance Optimization, Optimal Weights Problem

"Amy" <aburgin74@gmail.com> wrote in message <n1gusr$7u$1@newscl01ah.mathworks.com>...
> Hi All,
>
> I have what is probably a very simple problem that has me scratching my head. I am a beginner MatLab user and am therefore following the Mathworks documentation very closely. However, for the following, I cannot figure out what is wrong.
>
> I have the following function:
>
> function [Weights] = myfun(w,sigmahat)
> Weights = @(w)w'*sigmahat*w;
>
> And then the following code:
>
> muhat = [0.0427 0.0015 0.0285];
> sigmahat = [0.0100 0.0018 0.0011; 0.0018 0.0109 0.0026; 0.0011 0.0026 0.0199];
> TargetReturn=0.0285;
> X0=zeros(3,1);
> Aeq = [muhat;ones(1,3)];
> Beq = [TargetReturn;1];
> A = [];
> B = [];
> LB = zeros(3,1);
> UB = ones(3,1);
>
> [OptimalWeights, OptimalVariance] = fmincon(Weights,X0,A,B,Aeq,Beq,LB,UB);
>
> I get the following errors when I run this:
>
> Error using Weights (line 3)
> Not enough input arguments.
>
> Error in Practice (line 12)
> [OptimalWeights, OptimalVariance] = fmincon(Weights,X0,A,B,Aeq,Beq,LB,UB);
>
> I copied this almost verbatim from the following as it is the exact same problem:
> http://www.mathworks.com/matlabcentral/newsreader/view_thread/271112
>
> I need any help that can be offered. When I google the errors, I do not receive helpful information.
> Thank you so much for your time.

muhat = [0.0427 0.0015 0.0285];
sigmahat = [0.0100 0.0018 0.0011; 0.0018 0.0109 0.0026; 0.0011 0.0026 0.0199];
TargetReturn=0.0285;
X0=zeros(3,1);
Aeq = [muhat;ones(1,3)];
Beq = [TargetReturn;1];
A = [];
B = [];
LB = zeros(3,1);
UB = ones(3,1);

[OptimalWeights, OptimalVariance] = fmincon(@(x)myfun(x,sigmahat),X0,A,B,Aeq,Beq,LB,UB);

function [weights]=myfun(w,sigmahat)
weights=w'*sigmahat*w;

By the way: MATLAB's lsqlin is better suited to solve your problem than fmincon.

Best wishes
Torsten.