COVID-19 #13 (Still Optional)

There was a big jump in the number of cases in China this week, all coming out of Hubei province.
**********************************************
This is a good time to learn about false negatives and false positives.
In doing hypothesis testing, the null hypothesis is usually that something is not present in the data.
In some contexts, this is called a negative. Then a rejection of the null is a positive. A false negative is then when the test delivers a negative that should have been a positive. A false positive is when the test delivers a positive that should have been a negative.
All medical tests are actually statistical hypothesis tests. There’s been suspicion for a couple of weeks that the current test for SARS-CoV-2, the virus behind the COVID-19 outbreak, is delivering too many false negatives, especially in the early stages of the illness. So, too many people are being sent home, then they get sick, and by the time they get back to the hospital it may be too late for them.
In macro, this comes up in trying to forecast business cycle turning points. We have lots of indicators for this, but all the indicators yield so many false negatives and false positives that they’re basically useless.
**********************************************
Anyway, for Hubei only, China is now reporting clinical diagnoses. This means a patient goes to the doctor, and if their symptoms match they are diagnosed. It’s not clear if the tests are still being used or not.

2. Effectively, this means the govt is giving doctors greater discretion to clinically diagnose patients. Previously, they could only confirm cases with the nucleic acid test kits. But a govt expert said recently these were only 30-40 percent accurate.

1. There's a lot of buzz on the astronomical jump in cases from Hubei on Wednesday -- 14,840 new confirmed cases, almost 10 x compared to a day earlier. New deaths rose to 242, more than double fr day b4. The govt has changed the diagnostic criteria used to confirm cases. More ..

— Sui-Lee Wee 黄瑞黎 (@suilee) February 13, 2020

You may want to click through that tweet and follow the thread, this is just part 1 of 8. So there was a big jump in the numbers, which are now up to 60K infected. Deaths are not affected by this jump, but continue to climb towards 1,500.
**********************************************
The New York Times radio show The Daily ran a podcast today entitled “Fear, Fury and the Coronavirus”. It juxtaposes China’s efforts to fight the outbreak with its central government’s efforts to contain public anger about how they’ve handled it.
**********************************************
The Epoch Times is a newspaper based in Taiwan. It is decidedly against China’s current government. Having said that, they may have an easier time getting first person reports from inside China. They have been interviewing managers of crematoriums in Wuhan:

… He’s been seeing about four to five times the usual workload.
…
Of the 127 bodies received by the crematorium on Feb. 3, eight were diagnosed with the virus, while 48 were suspected of having the illness, based on their death certificates …

Let’s back out some math.

• If 127 is four to five times normal, then normal is 25-32 per day.
• This suggests that 95 to 102 extra bodies every day are from COVID-19.
• But, if only 8 of the bodies were officially diagnosed with the disease, this suggests that death rates are 12-13 times higher than publicly announced.

Form your own opinion about how truthful any of that is, or what the margin of error on my simple calculations is.
Also:

The official at the first-mentioned funeral home said that about 60 percent of the bodies come from private homes, while 38 percent are transported from hospitals.

That tends to confirm the videos available on the internet of vans cruising neighborhoods.