Chat about Novel Coronavirus - 2019-nCoV - COVID-19

You are reading a single comment by @Howard and its replies. Click here to read the full conversation.

•

Howard

The main problem with the FT chart is not the chart design - it’s the very good chance that the underlying data is crazy fuzzy

•

deleted in reply to @Howard

Both deaths and reported cases, or just reported cases? Deaths seem likely to be a more accurate measure, but again relies on testing.
•

NotThamesWater in reply to @Howard

the underlying data is crazy fuzzy

The underlying assumptions appear to be bonkers.

Mortality rates extrapolated from infection rates extrapolated from actual deaths (assumed / guessed at being C19 related), being then used to derive infection rates.

Any model validation will have sensitivity analysis and caveats, but your sois-disant statistician viro-epedemiologist experts pueyo, campbell et al. either ignore this, or (and prepare yourself for a shock) not qualified to make the statements they make, or, at the very least, should be taken with a big pich of salt.

Disclaimer: I am not qualified in any of the above fields. I know enough to know that I don;t know much, and that deferring to people that do this for a living is likely a GoodIdea™