re: Daily COVID Updated as of 11/2/20 8:00 PM

Government workers can’t handle that.

It would probably be difficult for sites like Worldometers to adjust for as well.

It would be more accurate. But it’s something we should just figure is baked in and look at big pictures. We’re getting more real-time data than probably any virus in history.

True, they probably can't. And its too late now. But every lab computer system I've worked with you have to put in the date and time of collection. Its simply a different capture.

Maybe next time thy will adjust at the beginning.

quote:

---

Can you just stop lying

---

This is not the thread for this. If you think it's a "lie", explain how or tell me the truth.

This is the link to the US state-level tracker that I have shared via Google Drive. Chromdome's State Level COVID-19 Daily Tracker
This is the link to the COVID-19 Country tracker that I have shared on Google Drive. Chromdome's COVID-19 Daily Tracker

The source for the data is https://www.worldometers.info/coronavirus/
The source for the testing data is from https://covidtracking.com/data/

You morons are still trying to figure out fake numbers? Lol guess what shite won’t add up. It’s been entertaining watching y’all rationalize this and that. 2,000 people die from the shits in the US every year don’t forget to put those down as China flu ??

I'm sorry to have to ask for clarification but can I get some?? Thanks.

IHME model update 4-May

A ton of blame casting on other country's data, and a bunch of methodology changes...

quote:

---

Today we launch a major update to our COVID-19 estimation framework: a multi-stage hybrid model. This modeling approach involves estimating COVID-19 deaths and infections, as well as viral transmission, in multiple stages. It leverages a hybrid modeling approach through its statistical component (deaths model), a new component quantifying the rates at which individuals move from being susceptible to exposed, then infected, and then recovered (known as SEIR), and the existing microsimulation component that estimates hospitalizations. We have built this modeling platform to allow for regular data updates and to be flexible enough to incorporate new types of covariates as they become available. Last, by relating transmission parameters to predictions of key drivers of COVID-19 epidemic trends – temperature, the percentage of populations living in dense areas, testing per capita, and human mobility – this new modeling approach will allow for a more comprehensive examination of how COVID-19’s toll could unfold in the coming months, taking into account these underlying drivers.

---

Ridiculuous. They couldn't resolve their simpler model with observations. I don't think this will do any better.

quote:

---

This is particularly important as many locations ease or end prior distancing policies without having a clear sense of how these actions could potentially affect COVID-19 trajectories given current trends in testing and mobility, among others. With our new modeling framework, we aim to provide a venue through which different COVID-19 epidemic scenarios and responses can be explored by location.

---

Wishcasting at it's finest.

Yep, it’s encouraging. If we can just have a Tuesday that isn’t the huge jumps we’ve been seeing I’ll be a lot more optimistic. Obviously it’ll be higher, but a smaller bump would be good to see.

Check out Spain and Italy recently for more good news. Huge drops there recently.

quote:

---

A ton of blame casting on other country's data, and a bunch of methodology changes...

---

I’ve been critical of the model due to the misspecification of the parameters and the underlying distribution assumptions.

That being said, the director of IHME was interviewed on the 538 podcast a couple of weeks ago and was very open about the flaws in the model (e.g., not symmetrical, skewed) and that it was underestimating deaths moving forward. In addition, I think some of their general ideas were sound (deaths were the least unreliable so they started there) as well as there initial assumptions that were later shown wrong as a function of human behavior (e.g., hospitalizations paradoxically decrease per death as they approach their peak).

And given how much criticism they get, from both sides of the argument, as well as flat out misrepresentations (e.g., people incorrectly attribute the 2 million and 240 thousand death predictions to them) I can at least respect that they’ve put themselves out there and have continued.

Regardless, it appears that this update is much more aligned with reality and it’s much closer to this other model I’ve been following that has seems pretty reliable and accurate.

AI COVID Projections

By the way, did you see Kevin Hassett (WH economic advisor, who infamously predicted the Dow was heading to 36,000 right before the Dotcom crash) pushed back against some of the projections and said his “cubic” model (apparently just went into excel and clicked the polynomial graph) is predicting we’ll be at about 0 deaths per day by May 15th.