distributions3
distributions3 copied to clipboard
prodist() followup
In #83 @zeileis implemented a generic prodist()
. I have two followup items to discuss about the implementation:
- An informatively named alias for
prodist()
, possiblydistributional_estimates()
orextract_distributions()
or something along these lines. Currently I likedistributional_estimates()
the most but do not love it. - Documentation. In particular, my understanding is that
prodist()
extracts distributional estimates of various population estimands from model objects. I think it would be good to (a) distinguish between a distributional estimator of estimands like E[Y|X] and "the distribution of a data point" in the documentation, especially since this is something likely to confuse students, and (b) to clarify in the documentation for eachprodist()
method what exactly the estimand is. Currently it is very hard to track down what distributional estimators are estimating without reading the source for each method and having a solid grasp of the variouspredict()
andforecast()
functions used in the implementation.
Thanks for making this a separate issue. Two quick comments:
- Maybe
predict_distribution()
would be a more verbose description? Internally we make apredict()
call and we return adistribution
object. - These are all good points. Additionally, we have to be a bit careful to distinguish the documentation of the generic and the methods. For the lm/glm methods its is more straightforward to say what exactly is extracted while for the generic we should be a bit more vague because the models for which methods are added might have different properties.