Youyang Gu (covid19-projections.com) dunking on IHME

2,768 Views | 11 Replies | Last: 5 yr ago by BiochemAg97
Keegan99
How long do you want to ignore this user?
AG


Recall that the IHME model got quite a bit of media attention earlier this month when they breathlessly forecast 400,000+ fatalities by December.



HotardAg07
How long do you want to ignore this user?
AG
Their model is so clearly trash. Youyang Gu's model has been basically spot on.
Keegan99
How long do you want to ignore this user?
AG
The IHME model was all-but provably wrong when those headlines hit.

The declining CLI metrics and hospitalization tallies precluded the upswing in reported deaths IHME forecast over the next month or so.

A properly skeptical press would have inquired.
GAC06
How long do you want to ignore this user?
AG
Did their model remember to factor in the 23,000 dead in Texas by the end of August?
Fitch
How long do you want to ignore this user?
AG
He's pretty much been doing this since April. His website ranks his own predictions against other national forecasts & pretty well bags on IHME.
BadMoonRisin
How long do you want to ignore this user?
AG
Moving to F16 in 5.......4.......3........2......

because this
Quote:

A properly skeptical press would have inquired.

THANK YOU FOR YOUR ATTENTION TO THIS MATTER!
BiochemAg97
How long do you want to ignore this user?
AG
Fitch said:

He's pretty much been doing this since April. His website ranks his own predictions against other national forecasts & pretty well bags on IHME.
As he should. IHME projections have been terrible from the beginning. Of course their pretty graphs always looked less terrible because they constantly realigned their model to yesterdays data. Easy to look good when the first half of your graph is the actual data and there is only 1-2 days of data to compare your model to.
Keegan99
How long do you want to ignore this user?
AG
nortex97
How long do you want to ignore this user?
AG
It's a shame when actual reason/science/statistical analysis conflicts with power grabbing propaganda media.
Aston94
How long do you want to ignore this user?
AG
If anyone is using the University of Washington model and spreading it as fact they are clearly doing so to scare people and drive clicks to their site. There is no basis whatsoever for their numbers.
ORAggieFan
How long do you want to ignore this user?
But, according to IHME they have deemed themselves most accurate.

BigOil
How long do you want to ignore this user?
AG
according to our own biased assessment, we're number 1!!!
BiochemAg97
How long do you want to ignore this user?
AG
ORAggieFan said:

But, according to IHME they have deemed themselves most accurate.


Looking at the global chart, it looks like LANL and YYG are better.

Comparing themselves to 5 others and considering themself "one of the most accurate" by coming in about 3/6. LOL.

Oh, and given they are in the top half, let's pool everyone and say we are better than taking the average too, just so we come in 3/7. LOL

Refresh
Page 1 of 1
 
×
subscribe Verify your student status
See Subscription Benefits
Trial only available to users who have never subscribed or participated in a previous trial.