home *** CD-ROM | disk | FTP | other *** search
/ NetNews Usenet Archive 1992 #20 / NN_1992_20.iso / spool / sci / math / stat / 1891 < prev    next >
Encoding:
Text File  |  1992-09-15  |  1.1 KB  |  30 lines
  1. Newsgroups: sci.math.stat
  2. Path: sparky!uunet!munnari.oz.au!manuel!NewsWatcher!user
  3. From: dos105@rsphy2.anu.edu.au (Darren O'Shaughnessy)
  4. Subject: Re: Darts
  5. Message-ID: <dos105-160992120231@150.203.49.43>
  6. Followup-To: sci.math.stat
  7. Sender: news@newshost.anu.edu.au
  8. Organization: Australian National University
  9. References: <dos105-130992174724@150.203.49.43>,<dos105-150992153659@150.203.49.43> <00960A78.F915B0E0@uinpla.npl.uiuc.edu>
  10. Date: Wed, 16 Sep 92 02:00:20 GMT
  11. Lines: 17
  12.  
  13. In article <00960A78.F915B0E0@uinpla.npl.uiuc.edu>,
  14. tayloe@uinpla.npl.uiuc.edu (Rex Tayloe) wrote:
  15. >  
  16. >      Try a gaussian in X and 1 in Y. (not necessarily of the same width). I
  17. > have had a similar experience when trying to monte carlo particle trajectories.
  18. > 
  19. >                                             Good Luck,
  20. > 
  21.  
  22.   But surely if you treat the X and Y directions as independent, which is
  23. what you are suggesting, the 2D distribution will be somewhat square- or
  24. rectangular- shaped? From experience, the vertical and horizontal errors
  25. have to be (anti-)
  26. correlated to some degree.
  27.  
  28.       Signed,
  29.               Pat Rourke  (you may not have heard of me)
  30.