Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
1997
*January
*February
*March
*April
*May
*June
*July
*August
*September
*October
*November
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 1997

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

Search the Archive

Integration error with Chebyshev polynomials

  • To: mathgroup at smc.vnet.net
  • Subject: [mg6357] Integration error with Chebyshev polynomials
  • From: farmer at shire.math.columbia.edu (David Farmer)
  • Date: Thu, 13 Mar 1997 01:20:14 -0500 (EST)
  • Organization: Columbia University Center for Telecommunications Research
  • Sender: owner-wri-mathgroup at wolfram.com

The following error was found in Mathematica 2.2:

In[1]:= Integrate[E^x ChebyshevT[3,x]/Sqrt[1-x^2], {x,-1,1}]

Out[1]= 0

That answer is not correct.  I don't believe it can be evaluated
explicitly.  In any case, the answer is not 0.  Mma 2.2 can do the
numerical integration correctly:

In[2]:= NIntegrate[E^x ChebyshevT[3,x]/Sqrt[1-x^2], {x,-1,1}]

Out[2]= 0.0696442

Oddly enough, Mathematica knows that it cannot evaluate this equivalent
integral:

In[3]:= Integrate[E^Cos[x] Cos[3 x],{x,0,Pi}]

                   Cos[x]
Out[3]= Integrate[E       Cos[3 x], {x, 0, Pi}]

I am just posting this because I understand that people in a position to
correct the error read this newsgroup.  Maybe it has already been fixed in
version 3.0.

Dave



  • Prev by Date: Re: batch mode with notebooks
  • Next by Date: Re: Mathematica Mie Code
  • Previous by thread: Re: wrong divergence?!?
  • Next by thread: Re: Mathematica Mie Code