Bug!
- To: mathgroup at smc.vnet.net
- Subject: [mg43617] Bug!
- From: Konstantin L Kouptsov <kouptsov at wsu.edu>
- Date: Tue, 23 Sep 2003 04:02:44 -0400 (EDT)
- Organization: Washington State University
- Sender: owner-wri-mathgroup at wolfram.com
In version 5.0:
In[1]:= Integrate[Exp[-(x^2 + y^2)/2 - I (k x + l y)],
{x,-Infinity,Infinity}, {y, -Infinity, Infinity}]
Out[1]:= (2*Pi)/E^((1/2)*(k^2 + l^2))
However, by adding S in the denominator one gets the wrong answer:
In[2]:= Integrate[Exp[-(x^2 + y^2)/2/S - I (k x + l y)],
{x,-Infinity,Infinity}, {y, -Infinity, Infinity}]
Out[2]:= 0
The tail I (k x + l y) is important. Without it, Mathematica gives the
correct answer Sqrt[2*Pi*S].
Konstantin.