Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Number of values in Gaussian Normal Distribution


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Number of values in Gaussian Normal Distribution
Date   Mon, 5 Oct 2009 14:20:23 -0700 (PDT)

--- On Mon, 5/10/09, vmz@vol.net.mt <vmz@vol.net.mt> wrote:
> I am doing this ,because I am  trying to understanding
> a little better, the concept of Gaussian Normal Distribution,
> given that it is what confidence intervals,and hypothesis
> testing ,boils down to.
> I really do not think that it is a fruitless enterprise,to
> get your hands a little dirty on such a pivoting concept.

In general simulation is a wonderful method for exploring 
statistical concepts, but in this case you are probably
learning more about how computers work than about the normal 
distribution: you are dealing here with concepts like infinity 
and infinitely small differences between values, and computers 
can necessarily only approximate those. This doesn't matter 
normally (no pun intended) as the approximations are so good 
we won't notice, but at one point it will start to bite if you 
push your computer hard enough. This is likely to occur in the
type of simulation you are running, in which case your results 
will say more about the limits of computers to approximate 
very small numbers rather than the normal distribution.

-- Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------

Send instant messages to your online friends http://uk.messenger.yahoo.com 

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index