Diana had trouble posting this.
To everyone: Please only post plain ASCII text.
Statalist is set to block all other messages. See FAQ.
Thanks,
m.p.
This is a two part question.
First:
I would like to select a random sample of 100 records. Not a random
sample of a # percentage of the data in memory.
So I do the following commands on a database that has 31638 records.
�---------------------------
. count
31638
. sort CLAIMNBR
. set seed 123456789
. sample .316
(31538 observations deleted)
. count
100 /*I have 100 records because I know sample .316 would give me
100 records*/
--------------------------------
My first question: Is there a better way to do this?
My second question: If there is not a better way... how do I count the
number of records and save that number to programmatically to come up
with sample .316? Instead of typing .316
in the program?
Diana Sticker, MPH
Center for Research on Occupational and Environmental Toxicology, L606
Oregon Health & Science University
Portland Oregon 97201
Phone: (503) 494-2899
Fax: (503) 494-7397
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
