<>
So, William, as you can see from Kit`s reply, the -cond-s nest without a
hitch. If you want more on that, try NJC`s (and his coauthor`s) tutorial:
http://www.stata-journal.com/article.html?article=pr0016
The difference between one call to -cond- and several is that you have to
create the uniform random numbers as a permanent (could also be temp, but
would not make much of a difference here) variable, as you can see in Kit`s
third line. If you simply nested -cond-s and tossed -runiform- calls for
each one in there, these would deliver different values from the unit
interval and the whole thing would not work...
As for your -simulate- thing, look at -help simulate- at the bottom where
you will find a nice template. I have always thought of MC analysis as a
tool to investigate the behavior of an estimator where, say, the small
sample behavior is unknown. What you want from it amounts to a test whether
-runiform- is a good random number generator, because an obvious analytical
result for your dice thing is available...
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von Kit Baum
Gesendet: Samstag, 21. Februar 2009 16:52
An: [email protected]
Betreff: st: re: basic monte carlo simulation
<>
William said
Non-transitive dice would be something like:
* die A has sides {2,2,4,4,9,9},
* die B has sides {1,1,6,6,8,8}, and
* die C has sides {3,3,5,5,7,7}.
A is likely to be higher than B
B is likely to be higher than C
C is likely to be higher than A
--------------
clear
set obs 10000
g un1 = runiform()
g firstdice=cond(un1<`=1/3', 2, cond(un1<`=2/3', 4, 9 ))
prop firstdice
--------------
likewise for second and third.
Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html
*
* 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/
*
* 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/
