bnowok / synthpop

Generating Synthetic Versions of Sensitive Microdata for Statistical Disclosure Control
40 stars 8 forks source link

A reasonable default approach for dates #15

Open bokov opened 3 years ago

bokov commented 3 years ago

A lot of use cases for synthetic data involve dates (not time-to-event). It might save users time trying to diagnose error messages if ?syn explicitly said:

The following methods are the only ones that can simulate date columns without pre-processing: normrank and sample

...if that is in fact the case.

It might also be helpful to have a vignette that explicitly demonstrates simulation of data that has a subject ID column, a date column, and the remaining columns observations on those respective subject at various times.