The really good models come with an hour-long explanation and pages of explanatory notes. They include dozens of scenarios, with prediction intervals and cautious inference
-
-
Show this thread
-
Now, I'm a fan of transparency, but I suspect if you published the entirety of the government's models, you'd get people focused on the absolute worst scenario, ignoring most of the complexity Which would be bad
Show this thread
End of conversation
New conversation -
-
-
Daily numbers of new cases and epidemic curve with daily number of new infections and deaths by day that symptoms were first detected are more useful -- they will flag successful intervention
-
Can you tell how popular someone is by their Facebook friends count? Maybe, but it's a bit more nuanced. If we're going to report a single number, perhaps it should be number of people suspected of breaking social isolation orders? They're the reason we can't have nice things
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.