Duplicates drop with prioritization on one value of a variable

[Adrij Chakraborty]

Hi, I am new to Statalist, so let me know if I have missed out on any submission protocols that make it easier for people to understand and respond to my query.

I am trying to use duplicates drop to drop duplicates. I know that the 'duplicates drop' command keeps only the first observation from the group of duplicates. But here, I want to drop the other observations from the group of duplicates and retain the one particular observation that is not the first occurrence in its group. Is that possible with 'duplicates drop'?

I'm using Stata 18.

[Ken Chui]

You may create a variable to indicate that the case has no "11" in the variable, and then apply the duplicates drop after sorting. The cases with "11" will receive a "0" in the "not_11" indicator, and will be sorted to the top and thus retained:

gen not_11 = (erwstat != 11)
gsort persnr_siab begorig endorig not_11
duplicates drop persnr_siab begorig endorig, force

[Adrij Chakraborty]

Thanks Ken, it worked wonderfully.