the modelsummary table is removed as it doesn’t pass crans’s checks
For robust estimations (the default), the model is now internally
updated with robust = FALSE and iterlim = 0 so
that the gradient and the hessian of the structural model are
computed.
Interface to sandwich (estfun and
bread methods) and to nonnest2
(llcont method).
predict(object, newdata = data) where data
is the data frame used to fit the model now returns the same as
fitted(object). Bug fixed thanks to Achim Zeileis and
Rebekka Topp.prediction,
margins, and modelsummary).texreg method.