Estimate the true scale of the epidemic

[kmichael08]

Numbers are underestimated for several reasons.
We could include estimates on the actual numbers, charts with confidence thresholds etc.

We could create or use existing analysis on this problem.
It could take number of tests / days from the 100th case / population etc. into account.
Nice addition here would be to explain the decisions of the model.
For instance, explain why we think Italy has 10 times more cases than reported?