Censored predictors

[StaffanBetner]

In some cases one could have predictor variables that are censored. It could be handled as a missing data problem, with restrictions on the values (which depends on whether the data is right-, left- or interval-censored)

Some references: https://doi.org/10.1111/j.1468-2354.2007.00470.x https://doi.org/10.1007/s12561-013-9099-4 (a multiple imputation approach) https://doi.org/10.1037/a0034274 (also MI)

[paul-buerkner]

Thanks! I will take a look if this is something I can support in brms at some point.

[JackCaster]

I am curious to know if there is any progress on this topic. I just stumbled upon a problem that may benefit from censored predictors.