I found this in a Chinese report: The median incubation period was 3.0 days (range, 0 to 24.0 days)Mean Mr Mustard II wrote:And now the new 'stay indoors for 7 days if symptomatic' rule will add to the reporting delay and increase spread within households (those lacking East Wings) then beyond, among schools, offices and everywhere else. Though slowing admissions fractionally.Vortex2 wrote:So according to my model that means a 3 day lag : infection => reported case.clv101 wrote:They say 10-20x more cases than positive tests. Or 5-10,000 cases.
Seems a bit short to me.
Another report: The coronavirus’ median incubation period, the new paper says, appears to be about five days, with the vast majority of people developing symptoms within 12 days.
Yet another report: The median incubation period was estimated to be 5.1 days
So let's say 5 days incubation period before symptoms.
Shorter than I had thought but it will take the infected person time to recognise the problem, and more time to be processed by the health system.
In parallel, I have reduced my model infected/hidden ratio from 8.3 to a more modest 6 based on recent data from China.
Using this lower number my model suggests that for the government's 10,000 currently infected figure to work you need a maximum infected->certified infected time of 4 to 5 days.
So the infected person will need to report and be classified by HMG within one day of symptoms appearing, for their data to even vaguely work.
A more realistic 5 days would give us a current infected figure of 12k - 16k ... and that is based on the most optimistic/generous data I can assemble.
Conclusion: HMG's estimate of 5k - 10k currently infected is very optimistic. 12k -16k is the minimum my model suggests, under optimistic constraints. More probable is a current infected figure of around 30k to 50k.
