Issue 48: Ebola vignette

[athowes]

Description

This PR will close #48. The point of this vignette is to demonstrate some of the more advanced capabilities of the epidist package.

Checklist

• My PR is based on a package issue and I have explicitly linked it.
• I have included the target issue or issues in the PR title in the for Issue(s) issue-numbers: PR title
• I have read the contribution guidelines.
• I have tested my changes locally.
• I have added or updated unit tests where necessary.
• I have updated the documentation if required.
• My code follows the established coding standards.
• I have added a news item linked to this PR.
• I have reviewed CI checks for this PR and addressed them as far as I am able.

[athowes]

Did you ever fit different delay distributions by sex or other covariates @seabbs @parksw3 for the ebola data?

That seems like a fun thing to do and would demonstrate some of the functionality of having this brms set-up.

[parksw3]

We did not. And I agree it would be a really useful demonstration. Currently packing to get ready to move to Canada tomorrow... will engage more after I move...

[seabbs]

Agree with both of you that this would be a nice thing to at least plan to do here as its within the functionality of the current package but hasn't been explored but is suggested as a good idea in Charniga et al. I say plan as we need to make sure any current design decisions don't close the door on this but if time also just doing it now would be good.

[athowes]

Note to self that after working on the in depth model vignette (#66) I am now in more of a position to implement a basic version of the gamma delay into this PR.

[athowes]

Updates based on #70:

• Can now do gamma
• Can now do formula for by space or sex more easily
• Need to update otherwise to be compatible with new function name

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 46.87%. Comparing base (fa1be26) to head (5ab998b).
Report is 61 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff             @@
##             main      #52       +/-   ##
===========================================
+ Coverage   28.49%   46.87%   +18.37%     
===========================================
  Files          10       12        +2     
  Lines         565      576       +11     
===========================================
+ Hits          161      270      +109     
+ Misses        404      306       -98     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[athowes]

Question... does this line list capture all known cases or some kind of subsample? Maybe good to say that early on esp alongside the epi curve for context (also to understand how representative the data are)

I think it's all confirmed and suspected cases from a scan of the paper but not 100% sure.

[athowes]

I've incorporated most of your feedback into the vignette. The main thing that helped was just switching to HMC from Laplace (solved the modelling issues). I have also created follow-up issues for:

• Extensions for Ebola vignette #285 Extensions to the vignette
• Ebola vignette clean-up #290 An additional round of clean-up

If this is passing and reasonable, I'd suggest merging this, then I can clean things up in a PR resolving #290. If you have more comments about it then we could put them in there. Reason for this strategy is that this thread has got overwhelming and so I'd like to get the 95% that is done banked then be in a clearer place to resolve any remaining things.

[seabbs]

LGTM to me. Happy that all additional work has been captured in new issues. This reads very nicely as is and I think surfaces some exciting functionality.