COVID-19 Statistical Prediction Model
- Thread starter BoilerBulldog
- Start date
I would have to understand the context around this statement. Did he say "If we don't start taking steps to mitigate the impact on our hospitals by slowing the rate of infections it is possible for there to be 2M Americans die", or did he say "OMG 2M Americans will die and this is the end of the world as we know it"? If it is the first, I think it is a reasonable statement. The second deserves some criticism.
I am not sure I have opined on this. I think it is fair to criticize the sitting president for giving out false information in a time of crisis, especially when he is overriding the experts standing next to him in an effort to cast himself in a better light. I have no problem with being positive. In fact I think that there is a time and place for our leaders (local/state/national, business/politics/religious/civic) to remain calm and reassure us all. I think Holcomb has done a pretty decent job in this vein. Of course my definition of positive does not include falsehoods and self-aggrandizing.
I am not sure I understand much of the public discourse in this country anymore.
If I can find the Tweet I will try to link it, but I am not adept at Twitter. I think Trump's transparency efforts around this are the right thing to do. He just needs to kick those meetings off then shut up and let the experts handle the meetings. I am not a big fan of Pence. As I've noted on here, Pence's lawyer went out his way to try to ruin my career, but having said that, I think Pence, except for sucking up to Trump, has handled those pressers pretty well. No question the public discourse is totally f'd up imho.
What about California? I missed it. FWIW this guy still thinks we are 4 days away from our peak.
For cases but not deaths right? His model seems more susceptible to higher density areas but when cases jump and deaths do not, it could also just be testing catching up. Maybe the model needs to be tweaked but at the end of the day it's the death spikes that worry me the most.
Sadly it's probably more than four days away. UMich has a team of analysts that have been pretty accurate. They think peak is still 2 weeks out at least.What about California? I missed it. FWIW this guy still thinks we are 4 days away from our peak.
A new facebook group for our favorite mathemetician!
https://www.facebook.com/groups/158015618707622/?fref=nf
https://www.facebook.com/groups/158015618707622/?fref=nf
Where can I go to place a SIZEABLE bet against this model ?
Peak in March? Before we even have testing kits and the chemicals to evaluate them? Death rate of 1400? Has he been watching the numbers recently? At the present rate, we will surpass this in days.
Sadly, the death rate won't peak til we get closer to the date where we get more ventilators - Ford said May. What am I missing?
Peak in March? Before we even have testing kits and the chemicals to evaluate them? Death rate of 1400? Has he been watching the numbers recently? At the present rate, we will surpass this in days.
Sadly, the death rate won't peak til we get closer to the date where we get more ventilators - Ford said May. What am I missing?
https://www.dailywire.com/news/epid...Mhty21Iu6KwIAPFbDQtcYL-vB4Y7k-zplIP3pFXUH7Iqo
Some good news here.
Some good news here.
He's updated his death numbers to reflect NYC:
And here's his explanation:
Covid-19 "USA Deaths Tracking with NY"
This graph is really revealing. And more importantly, it confirms an hypothesis we had a couple days ago. Cha-ching!
I have changed nothing about any of the models. Think of this as merely "accounting." This is the first time I have ever looked at this plot, because I just made it. I've been stewing over the right way to do this for a couple days, and it was the first thought on my mind when I woke up. (The power of the subconscious mind...)
Let me walk you through it. I usually don't like putting so many traces on a single graph for public viewing. It scares people off. But it is worth it, trust me. Take your time, digest each plot. Look at the scale for it, and don't move on until you get it intuitively.
Let's start with the axes. The left vertical axis is for all "Deaths/day" curves, the right vertical axis is for the two "sigmoids," or total death curves that finish into it. (Near 1400 and 2900.)
Let's start with the red dashed plot with the red dots, USA Reported deaths per day. The scale for this is in red, on the left. It is real data, so it is noisy. But it is reality. (Always start with the DATA!!)
The pink shaded line that is tracking the red-dots-curve is the sum of my original projection for the whole USA (deaths/day) plus the model predictions for NY (deaths/day). This "sum of models" is closely tracking what we are **actually** observing. (It is the sum of the light gray and blue peaks.)
Next, the solid dark gray line is my original model death tally estimate for the whole country (finishing on the right, near 1400 total death tally). The light gray peak is the cases/day for that model (use left scale).
The light blue curve is the peak corresponding to the deaths/day predicted by our NY model.
When you add the original model to the NY model, you get the dark red plot, finishing just under 3,000 total deaths.
The dark red dots and dashed curve is the total reported death tally for the entire US (actual data). Note the "inflection point" in the curve corresponding to when the NY death peak starts growing and the trace begins diverging from my earlier estimate. (This is exactly what a "secondary infection looks like in other countries, eg Italy.)
I am not saying NY is a secondary infection. I'm saying, when you consider the data and models separately *and* together, this approach provides valuable insight into what is going on in our country.
I will update this graph every day for a while. I might simplify it too... but there is so much good information in here, I don't want to leave anything out. I will think on it.
What I love about this, is that we hypothesized a couple of days ago that this way of thinking about the country would match the data... AND IT DOES. An a priori hypothesis confirmed, making us more confident in our models. This is the scientific method.
It also reveals that my original assumptions (and Midwestern bias?) about our country were likely incorrect, and that I will need to revise my assumptions. Note that I am being careful to let the data guide my thinking. Make an hypothesis, then test it. Learn. The scientific method works.
And here's his explanation:
Covid-19 "USA Deaths Tracking with NY"
This graph is really revealing. And more importantly, it confirms an hypothesis we had a couple days ago. Cha-ching!
I have changed nothing about any of the models. Think of this as merely "accounting." This is the first time I have ever looked at this plot, because I just made it. I've been stewing over the right way to do this for a couple days, and it was the first thought on my mind when I woke up. (The power of the subconscious mind...)
Let me walk you through it. I usually don't like putting so many traces on a single graph for public viewing. It scares people off. But it is worth it, trust me. Take your time, digest each plot. Look at the scale for it, and don't move on until you get it intuitively.
Let's start with the axes. The left vertical axis is for all "Deaths/day" curves, the right vertical axis is for the two "sigmoids," or total death curves that finish into it. (Near 1400 and 2900.)
Let's start with the red dashed plot with the red dots, USA Reported deaths per day. The scale for this is in red, on the left. It is real data, so it is noisy. But it is reality. (Always start with the DATA!!)
The pink shaded line that is tracking the red-dots-curve is the sum of my original projection for the whole USA (deaths/day) plus the model predictions for NY (deaths/day). This "sum of models" is closely tracking what we are **actually** observing. (It is the sum of the light gray and blue peaks.)
Next, the solid dark gray line is my original model death tally estimate for the whole country (finishing on the right, near 1400 total death tally). The light gray peak is the cases/day for that model (use left scale).
The light blue curve is the peak corresponding to the deaths/day predicted by our NY model.
When you add the original model to the NY model, you get the dark red plot, finishing just under 3,000 total deaths.
The dark red dots and dashed curve is the total reported death tally for the entire US (actual data). Note the "inflection point" in the curve corresponding to when the NY death peak starts growing and the trace begins diverging from my earlier estimate. (This is exactly what a "secondary infection looks like in other countries, eg Italy.)
I am not saying NY is a secondary infection. I'm saying, when you consider the data and models separately *and* together, this approach provides valuable insight into what is going on in our country.
I will update this graph every day for a while. I might simplify it too... but there is so much good information in here, I don't want to leave anything out. I will think on it.
What I love about this, is that we hypothesized a couple of days ago that this way of thinking about the country would match the data... AND IT DOES. An a priori hypothesis confirmed, making us more confident in our models. This is the scientific method.
It also reveals that my original assumptions (and Midwestern bias?) about our country were likely incorrect, and that I will need to revise my assumptions. Note that I am being careful to let the data guide my thinking. Make an hypothesis, then test it. Learn. The scientific method works.
Time will tell I guess. If he has a model that is wrong, he's accountable to it and cares more about trying to understand why. I appreciate that far more than what I'm seeing everywhere else. There have been numerous examples where these short term spikes turned back towards the model and are most likely due to testing/reporting issues vs the model tracking reality. There is also NYC and (maybe) Michigan where this spread a lot more than expected early on during the critical period.
Important to note that there absolutely will be secondary infections, especially after lockdowns cease. His models aren't projecting secondary infections, but the S Korea model shows what it will look like.
More than 3-4k will die from Covid-19. At the same time, this won't become 100k+ either.
More than 3-4k will die from Covid-19. At the same time, this won't become 100k+ either.
I sincerely hope you are right. That said, I m expecting 5,000+ deaths by the end of April and probably 500,000 cases
So is this “two weeks” I am paying attention to? Because we’ve had about a 6 weeks of “in two weeks.”
Search stories from a month ago and tell me what the predictions were. At some point the “in two weeks” has to be here. And if it doesn’t happen you have to let the guy in my neighborhood try to salvage his business and family.
I can find someone saying anything I want on the internet.
Which particular prediction(s) from 6 weeks ago do you think have turned out wrong?
OK. I furloughed or fired 75% of my workforce last week, the other 25% on reduced hours or salary. So... pack sand, I guess, with that last sentence.
Yes. It’s not hard to find any number of experts saying, “we will look like Italy in two weeks.” Well, more than two weeks ago.
And do you know why all this predictive modeling doesn’t work? Because computers don’t use common sense. I know people from China. People who were in China in early January. This didn’t start in early January in China and the first case didn’t happen in Chicago in late January.
Thousands of Chinese nationalists returned home from semester break in Chicago. Thousands of Chinese traveled to Chicago to visit their families in the south loop area and Argyle neighborhood. And somehow, somehow, never a single person brought this with or infected a neighbor? A co-worker? A stranger on the CTA. That is a pretty illogical scenario to believe.
These models all omit the existing communal immunity. And the current IDPH stats don’t report out where we should be if predictive models were right.
Interest comments by Gov. Cuomo today. “Maybe locking up everyone in their house and shutting down our business and closing our schools may not have been the best public health policy.”
Kinda think you need to get over the "two week" thing. It's altogether possible we'll look like Italy at some point, whether that's two weeks or six doesn't really matter that much, does it?
NYC is in a bad way.
Okay then. I am glad for you that you are smarter than the epidemiologists.
Dealing with the part of that which can be responded to with data.
On Feb 29 Italy had about 1100 cases.
2 weeks later on March 14 the US had 2300 confirmed cases on limited testing.
On March 9 Italy had 9000 cases and 500 dead.
Today the US has 85000 cases and 1300 dead.
Italy’s population is about 1/5th of the US
Predictions that we will look like Italy in 2 weeks have certainly not been clearly wrong. Their fatality rate has been a horror we have so far been spared.
In what tragically counts as hopeful, Italy’s daily death toll has stopped growing exponentially. I hope we do at least that well.
Good news. You were too high (but not by much).
Disappointed we didn't increase our testing volume (101k today va 113k yesterday). Hope we can get the point of care testing going soon.
A detail about that Tweet that bothers me to the point of British understatement.
That deaths chart is using a logarithmic scale while hiding the y axis. That’s just not cricket.
What a buffoon. Just admit you’re clueless and stop playing politics.
Computer modeling for Great Britain from mid-feb up to last week predicted 250-500k deaths. That number a few days ago was downward adjusted to 20k. Not by 20k. To 20k. What happened?
You also admit that testing has been limited. I expect that we have 3x to 10x as many cases, or have had. Which mean nobody knows the denominator so nobody knows the true mortality rate. The medical community is openly telling people with mild cases to not get tested. And there’s a large group of a symptomatic victims. So all these survivors are never accounted for. But the models are based on Italy which as Britain admitted is an anomaly.
So I’ll ask you one more time. Given the demographics of Chicago and the international visitors and residents. Do you believe that the first case of Covid 19 in Chicago occurred on January 24th as reported?
The only way the US looks like Italy is if deaths from all causes occurring in facilities treating COVID-19 patients are attributed to the virus. That's why Italy's death rate is an outlier. I've posted links to credible news sources over here including quotes from a prominent Italian medical professional who said 88% of the deaths attributed to CV there occurred in patients with multiple co-morbidities who would have died anyway within the next year.
I feel statistics are being reported have more to them. Just like a high scoring basketball player shooting a low percentage. USA is testing more people and has a higher population over more area than most countries.
Man, the guy is still sticking to his guns as of last night calling 3/30 the projected peak. Would like to see his US graph today. I think he’ll have to adjust it.
NYC is looking like Italy from two weeks ago.
Maybe the steps taken in other areas slowed the spread?
Yeah thankfully it didn't get that high. Hard to tell with the sheer lack of testing we have had but I am hoping that it just simply spread more than we think it did. That way we already have a lot of people out there presumably immune to it. I mentioned elsewhere but I know someone that I am pretty sure had it in early February. Had all the symptoms and it eventually progressed into pneumonia. He has since recovered thankfully but maybe there are others like him out there.
Talked to several of my MD friends yesterday and one this AM. They are very concerned about Covid.
I just talked to my brother who returned from Alabama yesterday (via RV) and said the differences by state and area were stark. One with beaches and thousands of people near each other; one which wore gloves and masks, gave him a pen in a plastic bag and didn't want it back; One still eating from open buffets and allowing people to touch the food & one which acted like nothing happened - talking face to face, shaking hands, etc... More cases are coming.
You have to click on the chart to see the full view on Twitter (because he's putting 2 different graphs on there). The left Y axis is labeled with units (although he doesn't call out log).
Here's S. Korea (with just one chart if you don't have twitter ... you can see the labels)
That’s a linear graph while the US graph is logarithmic. Just looking at the Korean graph and the US graph you might think they were growing similarly, but the Korean graph shows linear growth and the US graph shows exponential growth.
Hiding the Y axis is a cardinal data presentation sin.
I don’t think he did it on purpose. As you say, it’s artifact of how Twitter presented the images. It’s a really important detail though.
In his defense, he never put the Korean one side by side with the US (linear vs. log) .. and the Y axis is NOT hidden in any of his charts, just the way Twitter crops them when putting 2 charts in a single tweet.
