Error using symengine Unable to prove '2*(c1 + c2 + 1)^(1/2) == 0 & 35 < 5*c1 + 5*c2 + 5' literally.

I run the code and it returns error "Error using symengine Unable to prove '2*(c1 + c2 + 1)^(1/2) == 0 & 35 < 5*c1 + 5*c2 + 5' literally."

Please help me.

clear all
clc

syms c1 c2

S=50;
r=15;
mu=5;
sigma=2;
L=3;
m=5;

mu_ccLm=(c1+c2+(2*L)-m)*mu
sigma_ccLm=((c1+c2+(2*L)-m)^0.5)*sigma

mm=normcdf(S-r,mu_ccLm,sigma_ccLm)

Answer 1

[Walter Roberson]

syms c1 c2

S=50;
r=15;
mu=5;
sigma=2;
L=3;
m=5;

mu_ccLm=(c1+c2+(2*L)-m)*mu

mu_ccLm = 

sigma_ccLm=((c1+c2+(2*L)-m)^0.5)*sigma

sigma_ccLm = 

mm=normcdf(S-r,mu_ccLm,sigma_ccLm)

Conversion to logical from sym is not possible.

Error in normcdf>localnormcdf (line 127)
if sigma > 0

Error in normcdf (line 50)
[varargout{1:max(1,nargout)}] = localnormcdf(uflag,x,varargin{:});

That is the error message I expect for this situation.

There is no special symbolic normcdf() call, so your parameters are being processed by the normal numeric normcdf() call. Several of the statistics-related functions test input parameters numerically in order, such as if they want to have logic similar to

if sigma > 0
    do the normal calculation
else
    output 0, or nan, or as appropriate
end

There is not much you can do about this, other than to copy the implementing code for normcdf and editing out the numeric tests.