Creation of a loop without non-incremented numerical variables

[Michael Duarte Goncalves]

Dear Statalist Users,

This is my first time on Statalist, so I apologize in advance if this is not the place to ask my question. I have a very short question to ask. I have to work on a panel of data on graduate subjects with bachelors, masters, etc.

In order to isolate the public universities (let's call it publicU) from private ones, I tried to implement a code to remove only the public universities. However, when trying to implement a loop, the code does not work:

forval publicU= 309|328|334|335|339|354 {
gen new_`publicU'=1 if ts==`publicU'
replace new_`publicU'=0 if ts~=`publicU'
}

ts represents the codes that represent each university individually which interests me.

What would be a way to loop only on the above values, please?

Thank you very much in advance for your answer and help. Have a nice day.

Best regards,

Michael

[Maarten Buis]

So you have a variable ts that contains a numeric code for university, and you have a list of university codes that are public. Now you want to check if a value in ts is in your list (public) or not (private) and store the results in a new variable. You don't use loops for that in Stata. For checking is something is part of a list you can use the inlist() function, see help inlist . A minimal way of creating the variable you want is gen public = inlist(ts, 309, 328, 334, 335, 339, 354), with a bit more bells and whistles:

label define pub_lb 0 "private" 1 "public"
gen byte public:pub_lb = inlist(ts, 309, 328, 334, 335, 339, 354) if !missing(ts)
label var pub_lb "type of university"

[Michael Duarte Goncalves]

Hello Maarten,

Thank you very much for your answer.

Thank you also for the code and suggestions provided. This is exactly what I needed.

Best Regards,

--
Michael Duarte Gonçalves
Teaching and Research Assistant
Geneva School of Business Administration
(Haute École de Gestion de Genève - HES-SO Genève)
Rue de la Tambourine 17
CH-1227 Carouge
Switzerland