Closed nickreich closed 2 months ago
Now that some of the example data has changed to come from hubExamples
, I'm wondering if it makes more sense to change this running example to be using the forecast example from hubExamples
. Then maybe we can just say something at the beginning about the example data all being from that package, which contains subsets of data from real(?) forecasting exercises.
Thoughts @elray1 @eahowerton @nickreich?
I've toyed with this idea too @lshandross. A couple from me:
hubExamples
data in more detail earlier.hubExamples
. Also, I don't think we lose much because the initial example is just to explain the format of the model output class.Agree, sounds good to me!
Noting that there does not appear to be a clear introduction to the FluSight data that is being used as a running example. I think at least a few sentences, maybe an entire paragraph, is needed at some point fairly early on (either in section 3.1 or 4.1) to introduce the reader to the specifics about the model names, and the idea that these data are examples of real forecasts made by real models that have been excerpted for use as "example data" for this vignette.