Thanks Martin.
What if the dice were non-transitive? Is that where -simulate- comes
into play in order to generate the roll outcomes?
Say something like:
Dice 1: 1, 4, 4, 4, 4, 4
so 4 is 5/6 and 1 is 1/6
Rick
On Fri, Feb 20, 2009 at 3:17 PM, Martin Weiss <[email protected]> wrote:
> <>
> Normally, one would use -simulate- but this is more easily accomplished as
>
> ********* clear*
> set obs 10000
> g firstdice=1+int(6*runiform())
> g seconddice=1+int(6*runiform())
> g sumofdice=firstdice+seconddice
> *let`s see whether CIs conform to our idea of unbiased dice
> prop sum
> **********
>
> where you can edit the obs to the # of replications you want...
>
> HTH
> Martin _______________________
> ----- Original Message ----- From: "William Bishop" <[email protected]>
> To: <[email protected]>
> Sent: Saturday, February 21, 2009 12:08 AM
> Subject: st: basic monte carlo simulation
>
>
>> Trying to create a monte carlo simulation for throwing a pair of dice
>> with 6-sides, numbered 1-6 (each side equally likely, thus 1/6). So
>> there are 36 combinations of dice rolls and the sum will always be
>> between 2 (1 and 1) and 12 (6 and 6).
>>
>> With a large number of simulations, we should get the probabilities of:
>>
>> 2: 1/36 = 0.02778
>> 3: 2/36 = 0.05556
>> 4: 3/36 = 0.08333
>> 5: 4/36 = 0.11111
>> 6: 5/36 = 0.13889
>> 7: 6/36 = 0.16667
>> 8: 5/36 = 0.13889
>> 9: 4/36 = 0.11111
>> 10: 3/36 = 0.08333
>> 11: 2/36 = 0.05556
>> 12: 1/36 = 0.02778
>> *
>> * 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/
>
*
* 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/
