On Jan 19, 3:16 pm, jeff788 <[email protected]> wrote:
> I am trying to evaluate a definite integral using SAGE and am getting
> some errors.  I have evaluated the same integrals using Mathematica
> without a problem.  Here is the integral in question:
>
> integral(0.298321984000000/((13.1233333333333*X + 1)^2*
> (15.7466666666667*X + 1)),(X,0.05,0.3))
>
> I get the following error:
>
> Inverse of zero divisor?
>

This is because we use Maxima for our integration, and because we use
keepfloat:true in the "calculus copy" of Maxima.  Unfortunately, it
looks like somewhere along the line precision was lost and it thought
we divided by zero.  In Sage's Maxima:

(%i1) keepfloat: true;
(%o1)                                true
(%i2) integrate(0.298321984000000/((13.1233333333333*X + 1)^2*
(15.7466666666667*X + 1)),X,0.05,0.3);

Inverse of zero divisor?
 -- an error. To debug this try: debugmode(true);

Interestingly, if one doesn't do keepfloat:true, Maxima automatically
numerically evaluates this to be .003306490701348857




> I then tried using the numerical_integral command:
>
> numerical_integral(0.298321984000000/((13.1233333333333*X + 1)^2*
> (15.7466666666667*X + 1)),(X,0.05,0.3))
>
> for which I get he following error:
>
> TypeError: unable to simplify to float approximation

You used the wrong syntax (which we need to fix, however; perhaps it
is already in the rc of 4.3.1?).

sage: numerical_integral(0.298321984000000/((13.1233333333333*X + 1)
^2* (15.7466666666667*X + 1)),0.05,0.3)
(0.0033064907013495059, 3.6709421073472037e-17)

I hope this helps!  Thank you for your feedback.  Perhaps one of the
lurking Maxima devs on the list can tell us what the source of the
first error is, as I unfortunately don't understand all the details of
Maxima float-rational simplification.

But at any rate you should not need to use Mathematica for numerical
integration; we use the GSL at the level of a C library, I think, so
it should be fine for these purposes.

- kcrisman

- kcrisman
-- 
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org

Reply via email to