There is a ticket (sorry I don't have these on me, seems to be getting
a habit) for this with symbolic integration under something like
"Mathematics-style syntax for integrals".  It is not hard at all to
make (rectangular) regions do this with something nice.  The trickier
part is maintaining backward-compatibility for the deprecation period,
which is presumably why this has stalled.  I think that some of the
Hussein/Erocal patches under review with Pynac 0.11.0 (?) will make
this even easier than it already would have been.

Anyway, presumably something similar should be done with .nintegral
and numerical_integral (which should really be a method of Symbolic
Expressions, at least ones with one variable if the variable is given,
i.e. (x^2).numerical_integral(x,0,1) ).  It would also be really good
to make sure that numerical_integral can start handling cases with
parameters input at run time - am I correct that
numerical_integral(f(x,y).subs(x=3),...) still doesn't work that hot?
Or is that old info?


- kcrisman

On Feb 15, 2:37 pm, Robert Bradshaw <[email protected]>
wrote:
> On Feb 15, 2010, at 11:27 AM, Thomas Scofield wrote:
>
>
>
> > Can someone point me to examples/documentation on doing double  
> > integrals numerically?  That is, say you have to calculate a double  
> > integral with integrand f(x,y) for which no closed-form  
> > antiderivative (neither in x nor y) exists.  Assume my xy-region of  
> > integration is rectangular.  Is there a numerical_integral() type  
> > command for this case?
>
> Not that I know of, but you can nest them. For example, to do \int_3^4  
> \int_1^x sin(x^2/y) dy dx one can do
>
>      sage: numerical_integral(lambda x: numerical_integral(sin(x^2/y),  
> 1, x)[0], 3, 4)
>      (-0.78059401567023834, 8.6663344908170892e-15)
>
> Note that the [0] on the inner integral causes it to ignore the error  
> term there.
>
> - Robert

-- 
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