I had a first go at trying to reproduce the #2019nCoV #R0 results of Imai et al (Fig 1, Report 3 https://www.imperial.ac.uk/mrc-global-infectious-disease-analysis/news--wuhan-coronavirus/ …), using code from @JulRiou @C_Althaus and didn't have much luck. Can anyone see where I'm going wrong?https://gist.github.com/sdwfrost/b3d0c4cbff7ff460562162affceffb17 …
-
If you look at the report by Imai et al, their figure says that with a R0 of 2.6, you should get a median no of cases of 4000 by 18th Jan, and I get much higher. I thought I'd put it 'out there' so any causes for the inconsistency might be spotted (as I've tried and failed)
I just had a brief look and agree that there is inconsistency. Maybe it has something to do with the generation time distribution? It’s not clear whether Imai et al. use the SD as reported in Lipsitch et al., or maybe some other distribution.
-
-
Report does say on page 3 "for a negative binomial offspring distribution with k=0.64", but tweaking that doesn't fix results.
-
blegh. It's a bit too late SAST for clear thinking on this, but yeah - strange inconsistency with the conclusions vs assuming those params and generating results. Still, should be easy enough to consider various distributions for cases + times with the alternate code.
Koniec rozmowy
Nowa rozmowa -
Wydaje się, że ładowanie zajmuje dużo czasu.
Twitter jest przeciążony lub wystąpił chwilowy problem. Spróbuj ponownie lub sprawdź status Twittera, aby uzyskać więcej informacji.