Operands Not Scalar ?

[EDIT: 20110609 12:41 CDT - reformat - WDR]

function[x0]=Newton_Iteration(delta,epsilon,maxiter)

x0=5;

for iter=1:1:maxiter

  x1=x0-(f(x0)/df(x0));

  error=abs(x1-x0);

  relerror=2*error/(abs(x1) + delta);

  x0=x1;

  f_final=f(x0);

  if ((error<delta) || (relerror<delta) || (abs(f_final)<epsilon))
    break
  end
end

ERROR

Operands to the || and && operators must be convertible to logical scalar values.

Error in ==> Newton_Iteration at 17

  if ((error<delta) || (relerror<delta) || (abs(f_final)<epsilon))

Error in ==> RF_Model at 187
      T_e=Newton_Iteration(1e-10,1e-10,50);

What might be the problem...I am gettin this error once i change the / to ./ in a previous statement which was showing mldivider error...

pls help me out

[Sean de Wolski]

what is epsilon

size(delta)

size(error)

size(relerror)

size(f_final)

?

Also:

DON'T name your variable 'error' as this is a very useful builtin MATLAB function.

Answer 1

[Jan]

This is no valid Matlab syntax:

if ((error<delta) (relerror<delta) (abs(f_final)<epsilon))

I guess, you want something like this:

if (all(error_value < delta) && ...
    all(relerror < delta) && ...
    all(abs(f_final) < epsilon))
  break;
end

I've renamed your variable "error" to something, which does not equal the name of a toolbox function.

[Gowtham]

No actually it was supposed to be

if ((error_value<delta)|| (relerror<delta) ||(abs(f_final)<epsilon))

is it right now ?

[Sean de Wolski]

assuming all of those values are scalars.

[Gowtham]

hey thanks ..i got the working code...

Answer 2

[Walter Roberson]

You would encounter this problem if f(x0) or df(x0) return a non-scalar, which is something we cannot determine by looking at the code you show.