Skipping fzero error in a loop

I am trying to solve an algabraic equation with two parameters that are varying in a loop. The basic form of the program is as follows:

nn=linspace(0.01,1,10);
aa=logspace(0,3,10);
inn=length(nn);
iaa=length(aa);
ss=1+1./nn;

for ii=1:inn
    for jj=1:iaa
        ii
        jj
        Y0=[0 1];
        fun=@(Y) (1+Y)^ss(ii)-Y^ss(ii)-ss(ii)*aa(jj);
        YY(ii,jj)=fzero(fun,Y0)
    end
end

Now at some point at values of ii=2 & jj=7 the programs figures that a solution is not available, and stops. The error message reads: "The function values at the interval endpoints must differ in sign.", which understandably is due to the nature of the

Is there a way to get the program to keep running?

Answer 1

[Star Strider]

Yes!

Now I suppose you want to know how?

Define: ‘Y0=1’, not a vector of values. This gives you a matrix with many NaN values and many others with solutions outside ‘[0 1]’. You can convert these to NaN on your own later if you like, with:

YY(YY>1) = NaN;

The point is that the code completes and fills your matrix.

Answer 2

[the cyclist]

One possible solution is to use a try-catch control structure.