Number of ID Variable in Panel Data

[Mohsin Javed]

Hi, I have a panel data (large n and small T) with -id- as the panel variable and -year- as the time variable. The year variable ranges from 2000-2003 and id captures the number of firms.

The id variable is coded as follows:

--id-- --year--
1232 2000
1232 2001
1232 2002
1232 2003

1234 2000
1234 2001
1234 2002
1234 2003

When I summarise the dataset, minimum for id is 1 and max is 20,236. However, the id variable is not equally spaced so I cannot take 20,236 as the total number of firms in the data.

I need to find the number of firms in the dataset based on the id variable. How to proceed?

[Wouter Wakker]

codebook id

Or you can use the distinct command from SJ.

[Mohsin Javed]

Wouter Wakker: thanks, -distinct- command works.

I used the following website:

https://www.stata.com/support/faqs/d...-observations/

[Carlo Lazzaro]

Mohsin:
another approach relies on -egen- -tag- function:

. use "http://www.stata-press.com/data/r15/nlswork.dta"
(National Longitudinal Survey.  Young Women 14-26 years of age in 1968)

. egen flag=tag(idcode)

. tab flag

tag(idcode) |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |     23,823       83.49       83.49
          1 |      4,711       16.51      100.00
------------+-----------------------------------
      Total |     28,534      100.00

[Mohsin Javed]

Carlo Lazzaro: Thanks.