In the existing lit, the missingness is not on either side of the equation, because there is no equation, i.e. no predictive modeling. It simply assumes one has data, and it is up to the users what they want to do with the data after NAs have been dealt with, including predictive modeling. 🧵 1/
Comments
BTW, what specifically do you mean by "unbiased"? I'm not aware of any method to check this.
By unbiased I just mean what you’d get with the full data (no missingness in X) is the same (in expectation) as what you’d get if you estimate it using partially observed X
But again, the standard is to apply missingness methods to your data first, no "equation," then fit your model. My students and I are developing a package to facilitate this. 2/