Create an alternative variable for unique identifier in Stata

[Chuan Tang]

Hi All,

Recently I need to publish some dataset online. However, the unique identifier of my dataset contains some sensitive information that I am not expected to share. There are certain patterns inherent in the existing identifier variable that I do not want to share. I am wondering if Stata can help to create a new/alternative unique identifier based on existing identifier and can work to identify each observations (with duplicate observations)?

Any suggestions are welcome. Thanks!

[Clyde Schechter]

egen long new_identifier = group(existing_identifier)

[Nick Cox]

Clyde hits the nail on the head as usual. But note that this is an FAQ. The Statalist FAQ itself advises search of the Stata FAQs:

https://www.stata.com/support/faqs/d...p-identifiers/

[Chuan Tang]

Hi Clyde and Nick,

Thanks for your answers. It helps a lot. I will do a more comprehensive FAQ search first in the future.