Creating a new variable with all possible combinations of two existing variables

[Christina Chara]

Hi,

I would like to create a new variable that will have a unique values depending on the combination of two of my variables (i.e. I want a new variable at the intersection of my two existing variables). More specifically, I want to combine my Gender variable that is equal to 1 if male and 2 if female and I my Year of Labour Market entry variable that takes values from 1978 to 2011 into a new variable so I can then use this new variable for standard error clustering. I wanted to ask whether there is a quick way of telling Stata to do this rather than typing all possible combinations of the two variables and hence avoiding the risk of making a mistake.

Thanks in advance.

[Joseph Coveney]

If I read you correctly, then maybe something like

version 13.1

clear *
set more off

set obs `=2011 - 1978 + 1'
generate int year = 1978 - 1 + _n
expand = 2
bysort year: generate byte sex = _n

exit

[Clyde Schechter]

Joseph's code creates a data set with two variables representing all combinations of year (1978 through 2011) and sex. If Christina actually already has a data set containing the year and sex variables and wants to get a new variable that distinctly identifies each combination of them, then it would be

egen newvar = group(year sex)

[Christina Chara]

thanks a lot to both. I already have the year and sex variables in my data set so I will go with Clyde's suggestion and keep Joseph's suggestion in mind for the future.
Thanks again.