New to Mathematica, and trying to numerically evaluate two integrals:

  g[x_] := (x + 1) Log[x + 1] - x Log[x];
  C = NIntegrate[g[1/(Exp[1/x] - 1)], {x, 0, a}];
  P = NIntegrate[1/(x (Exp[1/x] - 1)), {x, 0, a}];

and plot C as a function P, if possible. Kindly help.

  • 2
    $\begingroup$ C is defined by Mathematica, look in the documentation. This works points=Table[{c=NIntegrate[g[1/(Exp[1/x]-1)], {x,0,a}], p=NIntegrate[1/(x (Exp[1/x]-1)), {x,0,a}]}, {a,0,1,.01}]; ListPlot[points, Joined->True] but gives some errors that you might be able to figure out. Otherwise NIntegrate demands every variable have been given some value first. $\endgroup$
    – Bill
    Commented Mar 17, 2015 at 22:38
  • $\begingroup$ By saying “plot C as a function P”, do you mean you want to use P as the independent variable of C? $\endgroup$
    – xzczd
    Commented Mar 18, 2015 at 7:30

1 Answer 1


According to your definition we have

g[x_] := (x + 1) Log[x + 1] - x Log[x]

The two functions performing the numerical integration are

fC[a_] := NIntegrate[g[1/(Exp[1/x] - 1)], {x, 0, a}]
fP[a_] := NIntegrate[1/(x/(Exp[1/x] - 1)), {x, 0, a}] 


1) variable names must begin with a lower case letter, hence I have written fC and fP instead of C and P, respectively.
2) delayed assignment (:=) is requested and the upper integration limit must appear as an argument of the function.

Now you could plot both functions, for instance like this

Plot[{fC[a], fP[a]}, {a, 0, 5}]
(* picture not shown here *)
  • $\begingroup$ I think you mistyped the definition of fP? $\endgroup$
    – xzczd
    Commented Mar 18, 2015 at 7:34
  • 1
    $\begingroup$ @xzczd: Yes, I did. Thanks for pointing it out. I've corrected it. $\endgroup$ Commented Mar 18, 2015 at 8:29
  • 1
    $\begingroup$ The OP wants to plot fC as a function fP, maybe like ParametricPlot[{fC[a],fP[a]},{a,0,5}]? $\endgroup$ Commented Mar 18, 2015 at 16:20
  • 1
    $\begingroup$ Yes, ParametricPlot.The resulting curve is almost a straigt line. $\endgroup$ Commented Mar 18, 2015 at 22:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.