Short answer — I didn’t start with one. I modeled based on raw figures, not proportions, and let the solver estimate the rates of movement between the three populations from that.

I used the number of recorded USA cases from early Feb through the second week of March as my initial data. Did the solving in R (the statistics software) with the deSolve library.

I haven’t had time to go back and update with latest numbers — going to all classes online takes a bit more time.

]]>At the moment models with floating points of inflection aren’t easily able to come up with a credible fit. A Gompertz curve (my preferrred for terrestrial poikilothermic vertebrates) still whips around like a firehose with just a few dozen cases added to a daily update, but overall there seems to be an indication with the passing of days that we’ll land somewhere between 20-40 thousand cases between the beginning and end of June, before we approach a plateau. Not great, but at least it’s not in the hundreds of thousands or the millions. Still, if our government had acted with greater alacrity the number might have otherwise been half of what will be realised.

I’ll try a Richard’s fit soon, to see if that can project further with greater accuracy.

One thing that has concerned me is that media commentators are quoting mortality by dividing the total number of deaths by the latest tally of infections. Given that a mortailty rate based on deaths realised to date should be predicated on the number of infections at the time that the deceased contracted their infections, the denominator should probably be the number of cases 7-10 days prior to the latest death count. On this basis Australia can expect 1-2% mortality – so somewhere between 200 to 800 dead based on the weekend’s rubbery projections. I’m just glad that it’s not the US’s death toll, which is already screaming upward past ten times that ball-park…

]]>https://www.sciencealert.com/the-new-coronavirus-could-have-been-percolating-innocently-in-humans-for-years

https://www.nature.com/articles/s41591-020-0820-9

As anyone who’s been following zoonoses knows, this is a very typical story. People tend to think – recall the initial days of HIV-AIDS, which turned out to be nothing but “initial” –– that these mutations appear out of nowhere and spread like wildfire, i.e., become pandemic diseases. That’s not the way it works.

]]>