transforming then imputing or vice-versa

[abdelilah arredouani]

Hi all,
I was wondering whether it is better to transform the data then impute the missing values or vice-versa. the literature is not helping because it is controversial.

Stata/IC 15.1 for Mac

[Maarten Buis]

I am flattered that you think that we more authoritative then the literature, but that is no justified. As far as I am aware, the debate hasn't settled yet. So you just have to make your own decision on which side you find most convincing.

[abdelilah arredouani]

Thanks Marteen. As i am not a statistician i am relying on opinions of experts in this list. i will follow the advice of this paper (transform then impute strategy)
https://onlinelibrary.wiley.com/doi/...1.2009.01215.x

stata 15.1 mac