add episodes on estimate severity

[avallecam]

This PR adds episode content for the middle task 3 (MT3)

I'll appreciate your specific feedback providing a clear next step to remove, change or add content.

some questions to guide feedback are:

• should we add how Nishiura's method differentiates from other cohort methods to estimate CFR?
• what other data sources/arguments can we use in challenges to increase the variability of countries shown?
• should we plan to add more episodes covering introductory concepts needed for {cfr}?

how to review?

• follow the steps in the reviewer role section of the contributing.md file

• Before running sandpaper::build_lesson(), in the config.yaml file deactivate these episodes:

episodes:
# - quantify-transmissibility.Rmd
# - create-forecast.Rmd
- severity-static.Rmd

• if local rendering doesn't work, please, report them in Issues.

• you can read online the knitted content of this PR in branch md-output-1 (drawback, figure captions not visible)

[avallecam]

@pratikunterwegs thank you for your feedback. I added your specific edit suggestions to most of them.

I have some open replies I would like to prioritize and request your re-review. Here are the links to each of them:

• explain the naive CFR concept
  • add episodes on estimate severity #1 (comment)
• on discrete distributions
  • add episodes on estimate severity #1 (comment)
• the theory behind the apparent low impact of delay-adjusted CFR at the end of the epidemic, compared to the ongoing trend
  • add episodes on estimate severity #1 (comment)
• on the misunderstanding sentence of the sampling uncertainty rule of thumb
  • add episodes on estimate severity #1 (comment)

Open to discussing any misunderstanding.

[pratikunterwegs]

Thanks @avallecam, this looks good to me. I've resolved the conversations as well, so feel free to merge once Carmen gives her fedback.

[avallecam]

feel free to merge once Carmen gives her feedback.

Thanks! I'll collect Carmen's feedback in a separate PR