MathGroup Archive 2007

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

Search the Archive

FindRoot with complex equations and variables?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg79093] FindRoot with complex equations and variables?
  • From: AES <siegman at stanford.edu>
  • Date: Wed, 18 Jul 2007 03:00:30 -0400 (EDT)
  • Organization: Stanford University

I have three eqns involving two complex variables w, u and one purely 
real variable g.  One equation is transcendental and complex (involves 
Bessel functions with the complex variables u, w in the arguments); one 
is complex-valued but algebraic (can easily be separated into real and 
imaginary parts); one is purely real (involves g and the real and 
imaginary parts of one of the complex variables).

Any thoughts on whether FindRoot is likely to be happier (i.e., function 
better) dealing with *three* equations (two complex, one real) involving 
the three variables w, u and g (including having to write Re[w] and 
Im[w] in the real equation), or with four equations (one complex, three 
purely real) involving wr, wi, ur, ui and g , all purely real?


  • Prev by Date: Coding an inverse permutation
  • Next by Date: Re: N-dimensional NIntegrate
  • Previous by thread: Re: Coding an inverse permutation
  • Next by thread: Re: Re: integration of piecewise convex bivariate function with 6 parameters