st: Confidence interval of a simulated experiment

[Najib Mozahem <najib.mozahem@gmail.com>]

Hi,
     I have a dataset in which different observations belong to different groups. I calculated a certain statistic. Next I wrote a small program that shuffles the observations between the different groups randomly and calculates and returns the same statistic. I call this program 1000 times using the simulate command. In order to see if the observed statistic (in the original data) is due to randomness i check the confidence intervals of the statistic which is returned by the simulation experiment. Should I refer to the percentiles which are displayed by the sum, det command or should I use the ci command?

Thanks
Najib 
*
*   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/