MathGroup Archive 2001

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

Search the Archive

Re: ComplexExpand & ExpIntegralEi

  • To: mathgroup at smc.vnet.net
  • Subject: [mg31845] Re: [mg31821] ComplexExpand & ExpIntegralEi
  • From: Andrzej Kozlowski <andrzej at tuins.ac.jp>
  • Date: Fri, 7 Dec 2001 05:56:27 -0500 (EST)
  • Sender: owner-wri-mathgroup at wolfram.com

You can avoid this problem by setting the option TargetFunctions to 
{Re,Im}:

In[2]:=
ComplexExpand[Conjugate[ExpIntegralEi[I x/a]],TargetFunctions->{Re,Im}]

Out[2]=
                      I x                         I x
I Im[ExpIntegralEi[-(---)]] + Re[ExpIntegralEi[-(---)]]
                       a                           a

It seems if ComplexExpand  is allowed to use Conjugate as a 
TargetFunction it will not remove it in cases like:

In[9]:=
ComplexExpand[F[Conjugate[x]]]

Out[9]=
F[Conjugate[x]]

although it will do so if F is replaced by many standard functions, e.g.:

In[11]:=
ComplexExpand[Sin[Conjugate[x]]]

Out[11]=
Sin[x]

but not

In[12]:=
ComplexExpand[ExpIntegralEi[Conjugate[x]]]
In[12]:=
ComplexExpand[ExpIntegralEi[Conjugate[x]]]

Out[12]=
I Im[ExpIntegralEi[Conjugate[x]]] +

   Re[ExpIntegralEi[Conjugate[x]]]


If you use TargetFunctions to "ban" Conjugate you will ge the expression 
you expected. It indeed looks like a bug to me.

Andrzej Kozlowski
Toyama International University
JAPAN
http://platon.c.u-tokyo.ac.jp/andrzej/

On Wednesday, December 5, 2001, at 08:51  PM, Blimbaum Jerry DLPC wrote:

> 	Why duzz  ComplexExpand[Conjugate[ExpIntegralEi[I x/a]]]  gives
> terms like  -I Conjugate[x]/Conjugate[a]   rather then -I x/a , where I 
> is
> the imaginary....Help says that ComplexExpand assumes all variables are
> real...
>
> 	thanks....jerry blimbaum   NSWC  Panama City, Fl
>
>
>



  • Prev by Date: Fit[] versus Regress[] inconsistent behahiour
  • Next by Date: Complicated Graph
  • Previous by thread: ComplexExpand & ExpIntegralEi
  • Next by thread: Matrix of possible combinations of N elements