[Date Index] [Thread Index] [Author Index]

Re: Please help with a Hypergeometric2F1 problem...

• To: mathgroup at smc.vnet.net
• Subject: [mg23430] Re: Please help with a Hypergeometric2F1 problem...
• From: "David Bailey" <db at salford-software.com>
• Date: Wed, 10 May 2000 02:32:15 -0400 (EDT)
• Organization: University of Salford, Salford, Manchester, UK
• References: <8e3b5h$kom@smc.vnet.net> <8eg5nd$gmg@smc.vnet.net>
• Sender: owner-wri-mathgroup at wolfram.com

Ronald Bruck <bruck at math.usc.edu> wrote in message
news:8eg5nd$gmg at smc.vnet.net...
>
> I am **impressed**.  While I am unable to coax Mathematica to
> differentiate this with respect to x and simplify the derivative to the
> original expression, when I plot the difference on [1,2] all I get is
> the typical roundoff noise.  It seems to be correct.

Yes indeed, if you do a series expansion about zero for both forms of the
integral the coefficients are identical (at least as far as I went) except
for the constant term, which is obviously arbitrary. I wonder, does this
mean that Mathematica does not 'know' how to translate this particular
hypergeometric, or is it that it does not 'simplify' hypergeometrics if the
result in terms of elementary functions is excessively complicated.

David Bailey
Salford Software