Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
2001
*January
*February
*March
*April
*May
*June
*July
*August
*September
*October
*November
*December
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 2001

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

Search the Archive

Integration problem / internal error ?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg31328] Integration problem / internal error ?
  • From: Madhusudan Singh <ar at sc.hc.de>
  • Date: Sun, 28 Oct 2001 03:07:08 -0500 (EST)
  • Organization: University of Michigan Electrical Engineering and Computer Science
  • Sender: owner-wri-mathgroup at wolfram.com

The problem is best stated as the following :

I have a monotonically decreasing function f[x+y] that is used in the
evaluation of a triple integral. The values of the function at the
endpoints are estimated to be 10^(-31) and 0.5 (taking max(x+y) and
min(x+y) over the integration interval).

Obviously the invocations of the function during the Monte Carlo
integration should lead to values intermediate between these.

However, an attempt to integrate leads to errors like "10^(-680) should be
a machine sized real number".

When f[z] is replaced with 1, or Chop[f[z],10^(-50)], the errors go
away.

I am puzzled. How can this possibly happen ? Is it an internal Mathematica
error ?




  • Prev by Date: Re: (-8)^(1/3)
  • Next by Date: Re: Plotting the histogram
  • Previous by thread: Chi Square test for goodness of fit
  • Next by thread: polynomials over finite fields