st: RE: year-wise rank function for multi-level data

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

As I understand it you want something like this: 

egen tag = tag(bg_code time) 
egen rank = rank(bg_sales) if tag, by(time) field 
bysort bg_code time (rank) : replace rank = rank[1] 

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

Apalak Khatua

I am using unbalanced panel data. There are multiple firms (firm_code)
within a business group (bg_code). I am considering sales for each
firm (i.e. firm_sales) as a proxy for firm size. So business group
size/sales (bg_sales) would be the summation of all firm-level sales
of that particular business group for that particular year (time).
Which I am calculating by:

egen bg_sales = total(firm_sales), by(bg_code time)

Now I would like to get the year-wise rank variable of all business
groups (bg_code) in terms of size (i.e. bg_sales) . Here the unique id
can be defined by bg_code and year (not firm_code). The rank variable
for the biggest business group (in terms of bg_sales or size of the
business group) for that particular year should be 1 and so on. How to
write the code to get this rank variable against each bg_code for each
year?


*
*   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/