Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.soft-sys.math.mathematica > #12616

R: Re: cauchy principal value double integral

From "Brambilla Roberto Luigi (RSE)" <Roberto.Brambilla@rse-web.it>
Newsgroups comp.soft-sys.math.mathematica
Subject R: Re: cauchy principal value double integral
Date 2012-12-05 08:09 +0000
Message-ID <k9mvgd$b7j$1@smc.vnet.net> (permalink)
References <k9hn9d$jg$1@smc.vnet.net> <20121204090930.187EF6881@smc.vnet.net>
Organization Time-Warner Telecom

Show all headers | View raw


Since the integrand is a positive quantity, obviously the integral has to be
 positive. Principal value has nothing to do with this integral since there
 is not the cancellation effect that requires a change in sign of the integer and. Mathematica 5.1 answers (correctly) with the non convergence of integra
l. Also Mathematica8 gives the same answer, with the option Principal Value-False.

I have divided the integration square in eight triangles by the two diagonals and consider the first one in the first quadrant: (0<x<1}&&(y,0,x)
And cut away a small e0 part of x near zero. The integral then becomes

R[a,e0]=8*Integrate[Integrate[1/(x^2 + y^2)^a, {y, 0, x}], {x, e0, 1}]

and performing the integrals (first the internal) is immediate to find that

R[a_,e0_]:=4 F[a](1-e0^(2(1-a))/(1-a)

where

F[a_]:= Hypergeometric2F1[1/2, a, 3/2, -1)

This last term results to be always positive 0<F[a]<1. So convergence/diverence of integral is determined by the other factor, i.e. convergence for e0-
>0 requires that a<1; For instance a=1/2
R[1/2,0]=8 ArcSAinh[1]

Cheers, Rob

-----Messaggio originale-----
Da: daniel.lichtblau0@gmail.com [mailto:daniel.lichtblau0@gmail.com] 
Inviato: marted=EC 4 dicembre 2012 10.10
A: mathgroup@smc.vnet.net
Oggetto: Re: cauchy principal value double integral

On Monday, December 3, 2012 2:18:53 AM UTC-6, Alex Krasnov wrote:
> I am unclear how to interpret the following result (Mathematica 8.0.4):
> 
> 
> 
> In:	Integrate[1/(x^2+y^2), {x, -1, 1}, {y, -1, 1}, PrincipalValue -> True]
> 
> Out:	-4*Catalan
> 
> 
> 
> How is Cauchy principal value defined in this case? Since the integrand is
 
> 
> circularly symmetric around (0,0), excluding a shrinking neighborhood 
> 
> around (0,0) is not useful. Perhaps this result is merely an issue with 
> 
> option handling for multiple integrals?
> 
> 
> 
> Alex

That does indeed look like a bug.

Daniel Lichtblau
Wolfram Research


RSE SpA ha adottato il Modello Organizzativo ai sensi del D.Lgs.231/2001, in
 forza del quale l'assunzione di obbligazioni da parte della Societ=E0 avvie
ne con firma di un procuratore, munito di idonei poteri.
RSE adopts a Compliance Programme under the Italian Law (D.Lgs.231/2001). Ac
cording to this RSE Compliance Programme, any commitment of RSE is taken by
 the signature of one Representative granted by a proper Power of Attorney.
 Le informazioni contenute in questo messaggio di posta elettronica sono ris
ervate e confidenziali e ne e' vietata la diffusione in qualsiasi modo o for
ma. Qualora Lei non fosse la persona destinataria del presente messaggio, La
 invitiamo a non diffonderlo e ad eliminarlo, dandone gentilmente comunicazi
one al mittente. The information included in this e-mail and any attachments
 are confidential and may also be privileged. If you are not the correct rec
ipient, you are kindly requested to notify the sender immediately, to cancel
 it and not to disclose the contents to any other person.

Back to comp.soft-sys.math.mathematica | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread


Thread

cauchy principal value double integral Alex Krasnov <akrasnov@eecs.berkeley.edu> - 2012-12-03 08:18 +0000
  Re: cauchy principal value double integral daniel.lichtblau0@gmail.com - 2012-12-04 09:08 +0000
    Re: cauchy principal value double integral Roland Franzius <roland.franzius@uos.de> - 2012-12-06 09:56 +0000
  R: Re: cauchy principal value double integral "Brambilla Roberto Luigi (RSE)" <Roberto.Brambilla@rse-web.it> - 2012-12-05 08:09 +0000
  Re: cauchy principal value double integral "Brambilla Roberto Luigi (RSE)" <Roberto.Brambilla@rse-web.it> - 2012-12-06 05:09 +0000

csiph-web