DOF adjustment for nested FE's in cluster about fixedeffectmodels.jl

[ptoche]

How come the reported number of observations is so different in the two cases? Can you post full reproducible examples with a commonly available dataset?

[jmboehm]

It's because reghdfe by default drops singletons before anything else, and does not report it in the table. It is the same data.

Sorry, I can't post an example from RDatasets because rdatasets on SSC seems to require Stata 14, which I don't have.

[matthieugomez]

It would be great to add this. Pull requests are welcome :)

…

On Sat, Dec 23, 2017 at 8:25 AM, Johannes Boehm ***@***.***> wrote: It's because reghdfe by default drops singletons before anything else, and does not report it in the table. It is the same data. Sorry, I can't post an example from RDatasets because rdatasets on SSC seems to require Stata 14, which I don't have. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <https://github.com/matthieugomez/FixedEffectModels.jl/issues/16#issuecomment-353726114>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AF733QHxFg0cjuW-gilCuF0CKVn-X02Mks5tDP9MgaJpZM4Q22vD> .

[jmboehm]

Did the update to dataframes 0.11 undo this? See diff here.

[matthieugomez]

Oops, not sure what happened. Just re-added it.
02d4635

[jmboehm]

Cool thanks a lot!