101  IDs

The way we deal with ID variables greatly depends on how they are created and what they represent. Id variables are typically integer-valued and denote the rows or observations in a data set.

While id variables are typically integers, it is common practice to parse them as strings. This means that you won’t run into any issues with rounding or overflow issues.

Typically, you don’t want to touch ID variables at all. If they are sequentially created, they act as a proxy for time, but if that is the case, you should have a proper predictor created that represents time more clearly.

If the ID variables are non-sequential, then there should be no information to gather from the variable at all. If you are having duplicates in your ID variable, then it can be a sign of an error.