fminbnd gives "Operands to the || and && operators must be convertible to logical scalar values"

I am trying to write value function iteration method. I write following function;

function val = valfnc(k)

global beta alpha kspace kmax kmin grid v0 k0 index

index = find(abs(k-kspace) <= (kmax-kmin)/(2*grid));
val = log((k0^alpha) - k) + beta*v0(1,index);
val = -val;

end 

The main code is

%initializations

clear all;
clc;

global alpha beta delta kmin kmax grid kspace k0 v0
alpha = 0.3;
beta = 0.95;
delta = 0.1;

kmin = 0;
kmax = 1;
grid = 1000;
kspace = kmin:(kmax-kmin)/grid:kmax;
[N,n] = size(kspace);

v0 = zeros(1,n);
v1 = zeros(1,n);


iteration = 1;

V = zeros(iteration, n);
policy = zeros(iteration, n);

fun = @valfnc;

 
while iteration < 11
    
    for i = 1:n
        k0 = kspace(1,i);
        
        kprime = fminbnd(fun,kmin,kmax);
        policy(iteration, i) = kprime;
        
        v1(1,i) = -valfnc(kprime); 
        V(iteration, i) = v1(1,i);
    end
    
    v0 = v1;
    iteration = iteration + 1;
end

Somehow I get the error message Operands to the || and && operators must be convertible to logical scalar values. Error in fminbnd (line 346) if ( (fu <= fw) || (w == xf)) Error in Hw1 (line 35) kprime = fminbnd(fun,kmin,kmax);

I have no idea about why this happens. I will be grateful for any comment.

Answer 1

[Cris LaPierre]

Your objective function must return a scalar value. It appears to not always be doing that. When iteration=1 and i=42, it returns a complex number:

>> fun(.5)

ans =

   2.1504 - 3.1416i

If I modify valfcn so that val = -abs(val), I no longer get the error.

Whether that is the correct approach or not is up to you.