Read on Twitter
A tale of two models: California Covid death projections from IHME and the Los Alamos National Laboratory team.
The difference highlights a key assumption about what is actually happening on the ground right now.
1/
The difference highlights a key assumption about what is actually happening on the ground right now.
1/
IHME model implicitly assumes that distancing measures are eventually successful in reducing new infections to zero, eventually drying up the "pipeline" for new deaths.
2/
2/
Los Alamos model assumes that new daily infections actually continue to ramp up and don't abate even into May, which continues to feed the deaths "pipeline":
3/
3/
As best I can tell, Los Alamos model assumes that growth in *confirmed* cases reflects *actual* growth in infections.
Which gets to the crux of it:
1. Is the infection still spreading at a high rate?
2. Or has it stabilized, but we're just measuring more of them?
4/
Which gets to the crux of it:
1. Is the infection still spreading at a high rate?
2. Or has it stabilized, but we're just measuring more of them?
4/
Los Alamos assumes #1, IHME model implies #2.
Not an epidemiologist, but it seems very likely the curve of actual infections should have a *very* different slope than the growth in confirmed cases.
But until we have better testing, we're still flying blind. So stay at home.
5/
Not an epidemiologist, but it seems very likely the curve of actual infections should have a *very* different slope than the growth in confirmed cases.
But until we have better testing, we're still flying blind. So stay at home.
5/
