creating a variable to identify the panel firms across several years

[Klodi Gjini]

Dear all, hope you are doing well. I am working with BEEPS data round IV-V and II-IV. I have appended the two datasets and I am facing some difficulties in xtset-ing because the unique identifier (idstd) is unique to the survey not across years. how can I create an id to identify the panel firms across several years? In my data I have identifier for specific years (seno2002 id2005 id2007 id2009). below you can find my code

* Example generated by -dataex-. To install: ssc install dataex
clear
input int(id2005 id2007) long id2009 int seno2002 long(panelid idstd) int year
. . 10082 . . 542185 2011
. . 10108 . . 542186 2011
. .   133 . . 542187 2011
. .     . . . 542188 2011
. .     . . . 542189 2011
. .     . . . 542190 2011
. .     . . . 542191 2011
. .     . . . 542192 2011
. .     . . . 542193 2011
. . 30081 . . 542194 2011
. .   999 . . 542195 2011
. . 20077 . . 542196 2011
. . 20083 . . 542197 2011
. . 20080 . . 542198 2011
. . 20085 . . 542199 2011
. . 10085 . . 542200 2011
. .     . . . 542201 2011
. .     . . . 542202 2011
. .   172 . . 542203 2011
. .  3621 . . 542204 2011
. . 30035 . . 542205 2011
. . 20025 . . 542206 2011
. .   188 . . 542207 2011
. .     . . . 542208 2011
. .  3624 . . 542209 2011
. . 20024 . . 542210 2011
. .   179 . . 542211 2011
. . 10025 . . 542212 2011
. .   200 . . 542213 2011
. .     . . . 542214 2011
. .     . . . 542215 2011
. .   163 . . 542216 2011
. .     . . . 542217 2011
. .   221 . . 542218 2011
. . 10030 . . 542219 2011
. .   199 . . 542220 2011
. .   237 . . 542221 2011
. .     . . . 542222 2011
. .     . . . 542223 2011
. .     . . . 542224 2011
. .     . . . 542225 2011
. .     . . . 542226 2011
. .     . . . 542227 2011
. .     . . . 542228 2011
. .     . . . 542229 2011
. .     . . . 542230 2011
. .     . . . 542231 2011
. .     . . . 542232 2011
. .     . . . 542233 2011
. .     . . . 542234 2011
. .     . . . 542235 2011
end

------------------ copy up to and including the previous line ------------------

thank you very much in advance

[Phil Bromiley]

You'll increase your chances of a helpful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

I don't understand. What firm identifier do you have that is the same for each dataset? If you don't have a common firm identifier, what possible way would you identify firms?

[Klodi Gjini]

Dear Phil, thank you for your answer. Maybe my question was not so clear. Panel dataset contains a variable “panel” that allows easy identification of panel observations. The idstd (unique id) variable can be linked to any of the underlying BEEPS data posted on the EBRD website and the Enterprise Surveys portal idstd means WEB STD FIRMID. But also panel dataset also contains the original country level panel ID variable that is matched across years (seno2002 id2005 id2007 id2009).. id 2007 means the id of the firms in 2007. id2005 means the id of the firms in 2005 and so on. so my question is: after appending the two BEEPS rounds ( II-IV and IV-V) i have several variables with id. I would like to create a new variable id_new id order to declare the panel. thank you very much in advance