New variables with conditions

[Marta Massera]

Hello ! I have a database where I need to create a new binary variable based on the values of 2 already existing variables. The 2 already existing variables that interests me are A (what is your profession) and B (what is the sector you're working in) because they help me identifying the professions of people interviewed. For the new variable I want to create, I want to select only people working on the transport sector starting from responses given in A and B. The new variable is going to be called Transport with 1 if they work on this sector and 0 if they don't.

In A : I have a subset of values that is certain that I want to assign 1 in my new variable
while I have another subset that I want to condition to the responses given in B.

Could the egen & cond() commands help me ? Thanks !

[Nick Cox]

Data example please. Please follow requests at https://www.statalist.org/forums/help#stata

Detail: cond() is a function, not a command.

[Maarten Buis]

I would do this in steps. I know I will make mistakes if I try to manage too many if conditions at the same time.

Lets say you want the new variable wanted to 1 if ( a is 4, 7, or 12) or ( ( a is 3 or 5) and b = 8 ) and 0 otherwise. As an additional requirement we want wanted to be missing when a is missing or (b is missing and (a is not 4, 7, or 12))

gen wanted = .
replace wanted = 1 if inlist(a,4,7,12)
replace wanted = 1 if inlist(a,3,5) & b==8
replace wanted = 0 if wanted == . & !missing(a,b)