Re: Mathematica computes a different answer in version 7
- To: mathgroup at smc.vnet.net
- Subject: [mg94028] Re: Mathematica computes a different answer in version 7
- From: mark mcclure <mcmcclur at unca.edu>
- Date: Tue, 2 Dec 2008 00:43:20 -0500 (EST)
- References: <gh0jl6$5mk$1@smc.vnet.net>
On Dec 1, 7:00 am, Jack Dyson <jackdyso... at libero.it> wrote:
> Consider this line:
> ...
> which is the moment convolution of two Gaussians. The answers I
> get in M6 and M7 are completely different!
> Is there something I am doing wrong here ?
The results are not equivalent, as you can tell by plugging in
random values for the parameters. Numerical integration
suggests that the V6 answer is correct while the V7 answer is
not.
This is not the first example I've seen like this. Here's one
I discovered in my complex variables class:
Integrate[Cos[m*x]/(1 + x^2), {x, -Infinity, Infinity},
Assumptions -> {Element[m, Integers], m > 0}]
In this example, it's fairly easy to use residue calculus to
show that the V6 answer is correct while the V7 answer is not.
I'd suggest you report this as a bug to support.wolfram.com.
Mark McClure
- Follow-Ups:
- Re: Manipulating a Binomial CDF
- From: Syd Geraghty <sydgeraghty@me.com>
- Manipulating a Binomial CDF
- From: Jamie Coventry <jcoventry@renaissance.edu.hk>
- Re: Manipulating a Binomial CDF