warn user if lm() produced coefficents with NA (ill-specified model) about broom

[alexpghayes]

Should broom should report a row of NA like in summary.lm in this case?

An argument for yes: tidy should let you probe the models you've fit, and this is certainly something you might want to know.

An argument for no: I can imagine cases were NAs could infect everything in an irritating way (i.e. you have a bunch of bootstrap models that occasionally drop enough data to NA out a coefficient and you want to calculate summaries across them).

(Aside: the plm model fits just fine for me and also drops the row of NAs)

[alexpghayes]

Dave and I talked about this earlier today. The consensus is that tidy will drop rows that contain only NA values. This is the current behavior.

I agree that modelling diagnostics should be way easier in R, it's just that broom isn't the place for them. Thanks for bringing this up!

[github-actions]

This issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.