st: RE: Random Sample Selection in Panel Data

[Nick Cox <n.j.cox@durham.ac.uk>]

One way to tackle this is that you perform sample selection on a dataset with one just one identifier per observation. Then you -merge- with the main dataset. 

Equivalently, tag just one observation per identifier, sample within that subset, and then expand to include all observations for each identifier. -egen, max()- is one way to do the expansion. 

In fact 

. search sample, faq  

shows that this is an FAQ, and that you could have identified relevant material directly within Stata, e.g. 

FAQ     . . . . . . . . . . . . . . . . . . Sampling clusters, not individuals
        . . . . . . . . . . . . . . . . . . . . . .  N. J. Cox and S. Merryman
        5/06    How can I sample clusters, not individuals?
                http://www.stata.com/support/faqs/data/sampleby.html

Nick 
n.j.cox@durham.ac.uk 

Dennis Kramer

I have a large panel data sets (4 years-- 250,000 + records per year)
and I want to generate four random sample groups to test the stability
of the estimates. However, I want to ensure that if a ID is selected
in Year 1 then are are subsequently selected into the sample random
sample for Years 2, 3, and 4.

I know for a cross-sectional random sampling the code is as follows:

generate rannum = uniform()
egen grp2 = cut(rannum), group(4)

Does anyone have any insight into modifying the above syntax to
automatically include years2, 3, 4, ids in the same sample as the
selected Year 1 ID??


*
*   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/