Kirk Borne
@KirkDBorne
Be careful with this one. Creating fake data and imputing missing values in your data can actually be a very BAD IDEA. If the intention is simply to demonstrate a pattern or trend in the overall general structure of a data set, then it may be okay. But if it’s done in order to claim that you now know the actual specific missing value, then that is misleading and incorrect. Missing data might itself be the actual signal of something more significant and actionable.
Source: https://t.co/3VLu6xS2ED
