Coronavirus 2019-nCoV

[Kung Wu]

Yes, my initial impression is that the data is obviously incorrect. Let's see, there are 30 million people in Texas, 21 million in Florida and, what, at least 7 in Arizona.

Nothing like excluding 20% of your data and calling it an 'outlier'....and how many other states where the virus is increasing are excluded? I heard there were about 13. Seens like Cali ought to be included in the 'outlier' group as well, given the fact that the virus is surging in the greater Los Angeles area.

Any good statistician would at least include it in a separate set. And it should be advertised on their website.

That is directly from the CDC.

I believe I have access to all the data I need to compile a US wide rate of hospitalizations. I'll know when I am back in town, but I will predict this: Including every state that reports hospitalizations will NOT prove your point and will make your point look weaker. Including them all will cause that graph to widen even further from the peak.

FL has been linear, for example, meaning no change up or down -- and it's considered by you to be a hot spot. Linear weekly hospitalizations will do nothing to the graph at all.

AZ hospitalizations are dropping from their small 5-day peak from three weeks ago. After that? Sharp decline. Including it will cause the graph to digress more, not increase.

TX is showing growth in hospitalizations, however not to the extent you think it is. The issue is that COUNTIES are hot spots, not states. The more counties you include, the more that graph is going to rise and fall. But we shall see when I am back.

That data set was specifically built because those counties were designated as "emerging" and/or existing hot spots. Adding in the 80% of the rest of the country -- the vast majority being rural -- is not going to strengthen your case.

[RoyalShock]

I did not see that for AZ. I saw that hospitalizations peeked three weeks ago and have fallen sharply since, and deaths peeked two weeks ago and have fallen sharply since. On vacation so not in a situation to share graphs for several days, but maybe someone else can show the AZ dashboard graph.

Capture.JPG

[Kung Wu]

Capture.JPG

That doesn't appear to be AZ's dashboard. Let me see of I can find it and compare.

[Kung Wu]

Yeah those daily death totals do not look like AZ's Department of Health data.

https://azdhs.gov/preparedness/epidemiology-disease-control/infectious-disease-epidemiology/covid-19/dashboards/index.php

Click on deaths and scroll down.

Maybe there is a bug in the tool you are using?

[RoyalShock]

Yeah those daily death totals do not look like AZ's Department of Health data.

https://azdhs.gov/preparedness/epidemiology-disease-control/infectious-disease-epidemiology/covid-19/dashboards/index.php

Click on deaths and scroll down.

Maybe there is a bug in the tool you are using?

Very well may be. I came across that site yesterday but didn't look into how they collect their data.

[WstateU]

Very well may be. I came across that site yesterday but didn't look into how they collect their data.

Monkey Math?

[SB Shock]

On the news, when they report the state and county caronavirus figures, they say the total cases, new cases, and the number who have recovered.

the question is not do they report it, I have spreadsheets of what they report. The question is how do they determine the actual number of who has recovered, and is it accurate?

[Shockm]

the question is not do they report it, I have spreadsheets of what they report. The question is how do they determine the actual number of who has recovered, and is it accurate?

My guess would be they test them multiple times, until they test negative (then report it). I'm not sure if they are telling the truth or not so they may not be accurate. Maybe I'm missing something else you are inferring.

[SB Shock]

My guess would be they test them multiple times, until they test negative (then report it). I'm not sure if they are telling the truth or not so they may not be accurate. Maybe I'm missing something else you are inferring.

They are not retesting people except maybe those who are sick enough hospital. There is not enough test kits to be retesting.

The state guidance is once you have symptoms or tested positive to either wait 10 days or 72 hours after symptoms stop without fever, whichever is longer. State is Only prioritizing the testing for people with complications or other public health priorities like health care workers.

[Aargh]

Looks like we can add Goldman-Sachs to the freedom-hating liberals who want everybody to wear masks.

https://www.forbes.com/sites/sarahha...w#2c70e59656f1

[wufan]

Looks like we can add Goldman-Sachs to the freedom-hating liberals who want everybody to wear masks.

https://www.forbes.com/sites/sarahha...w#2c70e59656f1

At least for me, it’s not about freedom. It’s about BS. I guess we can add Goldman-Sachs to the list of companies that are ignorant or like BS.

[wufan]

I case you were concerned that we had lots of positives and not enough deaths, A Yale study is here to help!

https://apple.news/A7UjwEWWwRbyoBzjCHLQr1g

Soon we will get the adjusted real numbers of Covid deaths.

[SB Shock]

They are not retesting people except maybe those who are sick enough hospital. There is not enough test kits to be retesting.

The state guidance is once you have symptoms or tested positive to either wait 10 days or 72 hours after symptoms stop without fever, whichever is longer. State is Only prioritizing the testing for people with complications or other public health priorities like health care workers.

ultimately what I’m inferring is I believe recovery data is not actually “data” but is a estimated number (which is at risk of being manipulated).

[wufan]

ultimately what I’m inferring is I believe recovery data is not actually “data” but is a estimated number (which is at risk of being manipulated).

As noted by the LA Times article, estimates are the only way to understand the full effect of the virus...which is Trumps fault.

[revenge_of_shocka_khan]

I haven't posted an update here lately.......so I will today.

If you are watching TV, it is obvious our hospitalizations and ICU patient census seems to be going up, as well as the number on ventilators. Deaths are flat (for now) if we don't start having a peak soon, I'm going to say we will probably see more people who get really sick this wave (and less who die outright). I'm going to say this is because of the discovery of effective therapies.

however

I'm going to hold off on making a definitive statement, as our public officials are telling us we will run out of hospital beds in about two weeks. We have around 20% available now, and the number in the hospital continues to grow, so this is a secondary worry, as if hospital bed availability disappears, I'm thinking people will be dying at home. Not only that, the people who have serious medical events (car wrecks, heart attacks and so forth) may not get treatment, so they will probably be lumped into some 'collateral damage' category.

All this will be avoidable if people continue to wear masks in public, practice social distancing and hopefully start bending the curve in the next two weeks.