Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Questions for random data generation and value label


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Questions for random data generation and value label
Date   Tue, 12 Mar 2013 18:04:00 +0100

On Tue, Mar 12, 2013 at 5:33 PM, Yu Xue <snowrain@gmail.com> wrote:
> Thanks everyone for answering my question, especially Joseph! I think
> you offered a right solution to my problem, although the result is
> still not accurate.

What do you mean with "not accurate"? I suspect it is accurate in the
sense that it does what it promisses to do, except that you wanted it
to do something different. So in order to answer your question we need
to know _exactly_ what you want.

This leads (again) to the question: What is the distribution from
which you want to draw your samples? In order to draw a random sample
you first need to define a distribution from which to draw that
sample. There is no other way and there can be no other way. For
example Joseph's solution does that by specifying some transformation
of a normal distribution. A mean, standard deviation and a range on
its own does not give you enough information to define a distribution
from which to draw a random sample.  In fact, specifying both the
range and the standard deviation will often makes things harder
(impossible). For some reason you are not willing to tell us what
distribution you want to use. One very reasoable response to that
would be: "OK, if you don't want to give the information necessary for
us to answer your question, than it's your problem. Good luck."

However, lets give it one more try. Maybe if you can tell us exactly
what you mean with "not accurate" we can deduce from that what it is
that you want to achieve. So tell us _exactly_ what you told Stata to
do in order to determine whether or not your result was "accurate",
what Stata told you in return, and why you think that those results
indicate that the sample is not accurate.

-- Maarten

---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

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


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