I’ve heard twice on the radio today that social distancing measures are now suggested til 2022 hahaha this won’t last much longer. But hey at least they are finally acknowledging that people are still gonna get this once they let us all back out.
Printable View
I’ve heard twice on the radio today that social distancing measures are now suggested til 2022 hahaha this won’t last much longer. But hey at least they are finally acknowledging that people are still gonna get this once they let us all back out.
Entirely plausible for Sweden. An updated set of numbers would be 114, 96, 106, 77, 17, 12, 20, 114. Is it increasing, or basically flat for 8 days? I don't know, but it doesn't look like exponential growth.
Yeah, Lazytroll37 has written pages and pages on why that number (3154) was wrong (and I'm sure it was wrong). If you look at the WHO data (which I'm sure I will be told is also wrong) from the last several days you see: 1895, 1925, 1931, 1920, 1528, 1504. That looks kind of flat or maybe even decreasing. Which was kind of my point from a few days ago - maybe we hit the peak already. I don't know; maybe we didn't. I never said we did, only that having hit the peak already would be the best case scenario. I thought that was kind of hard to argue, but a wall of words from Lazytroll37 showed otherwise. In any case, repeatedly claiming with authority that we haven't hit the peak yet, or we aren't in the thick of it, or whatever in the absence of the ability to predict the future is kind of silly for someone who teaches statistics.
I wonder if we'll all get a lecture on prediction models again?
I hope so! He's so much fun!
The so-called "prediction" models should be an embarrassment to those who built them. It boggles the mind how anyone could defend them.
The officially sanctioned IHME model is a joke. Set aside the fact its death projections were revised from 240K to 60K within a matter of days. Just look at its uncertainty bounds. I mean, how is it useful? For tomorrow, it's prediction for nationwide hospital bed demand ranges from 13K to 181K...right, let me plan across two orders of magnitude.
It's unrelated to the main topic of his post.
Personally, I think physical interaction cannot be banned any longer. For the time being, there need to be guidelines on sanitary practices (e.g. mask wearing, 6 feet of separation, hand washing at regular intervals, regular facility cleaning) which, if followed, will shield businesses from lawsuits by their customers and workers.
No lecture from me. But to illustrate the state of virus prediction models consider:
Predicting a roulette wheel. One way to go about it is to measure and track every physical parameter of the wheel and the rooms environment and solve all the dynamic equations. People have attempted this as a “cheat” with very limited success. It was based on the idea that some wheels run less true than others.
Or you can apply very basic probability theory and model a roulette wheel extremely accurately without knowing any physics or engineering. That’s how casinos calculate odds.
What you can’t do is apply a probability model to a complex problem if you don’t know the parameters.
We’ve destroyed the entire global economy based on a roulette wheel probability model without knowing the number of positions the ball can land in.
Came across this article today. The police are now treating working out in a gym as a crime. People are telling police they are just cleaning a gym instead of working out in order to avoid being arrested or fined. A few weeks ago this would seem like it was taken from a fictional sequel to Orwell's 1984, now it's a real news article. How crazy is this?
Coronavirus: Was Elmhurst Gym Open? | Elmhurst, IL Patch