I need a little help in data management

[Amit Nandan]

Dear users:
I need a little help. I have around 17000 names of firms. These are arranged like (1) A B G Motors Limited,(2) A B B Instrumentation Limited [Merged]. In my sample, we have companies that are unmerged & unmerged. The name ends with [Merged] at the end of merged companies. How to identify such companies in a sample of 17000 companies.

Thank you!
With kind regards,
Amit

[Mike Lacy]

If by "identify" you mean "create a separate binary variable to identify merged companies," you could do this:

gen merged = strpos(lower(YourCompanyNameVariable), "[merged") > 0

Various refinements to this (removing spaces etc.) are possible depending on whether your data is "clean," but without further information, this would be a reasonable first try.

[Hemanshu Kumar]

On the off-chance that "[Merged]" might appear elsewhere in the string and should then not count, you can use a simple regular expression match instead:

clear
input str30 name
"[my] [Merged] [Company]"
"Othercompany [Merged]"
end

gen wanted = ustrregexm(name,"\[Merged\]$")
li, noobs

  +----------------------------------+
  |                    name   wanted |
  |----------------------------------|
  | [my] [Merged] [Company]        0 |
  |   Othercompany [Merged]        1 |
  +----------------------------------+

[Nick Cox]

If a string ends in [Merged] then it's true that (assuming variable name name)

substr(name, -8, 8) == "{Merged]"

but all sorts of things could undermine the success of testing for that condition, including extra spaces, different use of upper and lower case, and other punctuation.

[Amit Nandan]

Thank you, professors!