How can I solve this myfun/numel problem?

function F = myfun(p)
s1 =(exp(5-p(1)))/(1+exp(5-p(1))+exp(3.5-p(2))+exp(5-p(3)));
s2 =(exp(3.5-p(2)))/(1+exp(5-p(1))+exp(3.5-p(2))+exp(5-p(3)));
s3 =(exp(5-p(3)))/(1+exp(5-p(1))+exp(3.5-p(2))+exp(5-p(3)));
F = [s1+(p(1)-1)*(-s1)*(1-s1)+(p(2)-1)*(s1*s2); 
  s2+(p(1)-1)*(s1*s2)+(p(2)-1)*(-s2)*(1-s2); 
  s3+(p(3)-1)*(-s3)*(1-s3)];

p0 = [-5; -5];  
options = optimoptions('fsolve','Display','iter'); 
[p,fval] = fsolve(@myfun,p0,options)

Attempted to access p(3); index out of bounds because numel(p)=2.

Error in myfun (line 2)
s1 =(exp(5-p(1)))/(1+exp(5-p(1))+exp(3.5-p(2))+exp(5-p(3)));

Error in fsolve (line 217)
          fuser = feval(funfcn{3},x,varargin{:});

Caused by:
  Failure in initial user-supplied objective function evaluation. FSOLVE cannot
  continue.

Answer 1

[Matt J]

You need to provide fsolve an initial guess for p(3)

Answer 2

[Image Analyst]

If p has 2 elements, you'll have to use ./ instead of /, like this:

s1 =(exp(5-p(1))) ./ (1+exp(5-p(1))+exp(3.5-p(2))+exp(5-p(3)));

Same for the other equations. For F, use .* (dot star) instead of * (star).