Author Topic: puzzled  (Read 42 times)

gsgs

  • Administrator
  • Full Member
  • *****
  • Posts: 196
    • View Profile
puzzled
« on: February 19, 2020, 08:08:00 pm »

puzzled
February 16th, 2020, 01:52 PM
I'm totally puzzled, what's going on.
It's obvious since ~Feb09 that cases in China are going down
and it becomes clearer every day.
Still flubies and experts and health officials and most reporters are ignoring this
and talk as if the pandemic thread were the same as in early Feb.
Why should Singapore or Japan or UK be a problem with just a few cases
when China could handle it with thousands of cases ?


February 16th, 2020, 02:29 PM
there are lots of papers that estimate the Chinese underreporting. Outside Hubei it
was a factor of 1.4 afair in a recent paper.
But it doesn't matter, we just observe the trend, not the total numbers.
Why should the underreporting have changed ? And in a way to produce
a stable downtrend ?
And no, you can't produce this deliberately. In all the provinces.
And the reporters are especially WHO do trust the numbers, reproduce them.
And those who talk about it do acknowledge that they are going down,
just that it's commonly ignored/avoided.

Well, some reporters say they are still going up, but they mean the cumulative
numbers which of course are increasing.


China ex Hubei ~1.5
Hubei ex Wuhan ~1.9
Wuhan~19 (Jan.29)

I read recently.

for the trend underreporting doesn't matter, unless they systematically change the level of underreporting over time


Statistical Inference for Coronavirus Infected Patients in Wuhan
Yong-Dao Zhou1, Jianghu (James) Dong2, *1
https://www.medrxiv.org/content/10.1...774v1.full.pdf



February 18th, 2020, 02:53 PM
you may remember Helen Branswell from our H5N1 discussions,
she writes (good) COVID articles now :
https://www.statnews.com/staff/helen-branswell/
Now I found that at twitter on Feb/09 she drew the chart of daily new cases in China and wrote:
> Chart of daily increases. Don't read too much into the decline; not sure what it means.

Well, it means no pandemic (yet).
On the same day I wrote in the German forum here :
let's get on our knees to praise this rare luck. While the wording was from a German
comedian, I was serious about it and it did express my feeling at that time.
The days after Jan.24 it had looked very bad to me .
So, why was Branswell so relaxed, indifferent ? She did seem to trust the Chinese numbers,
lists them frequently, discusses Chinese studies, saw many other epidemic charts before.
The numbers continued to go down but she didn't comment (afaics)
In her articles,tweets after Feb.9 she still sounded concerned, didn't mention the decline.

I'm puzzled.


February 18th, 2020, 03:17 PM
yesterday Zhong Nanshang estimates that the epidemic may stabilize by the end of April
He expects the peak in China by the end of February.
But that won't necessarily be the trend reversal.

He had earlier said on Feb02
I haven't looked at the data this morning
The number is rising now and the rising trend will continue for some time, but I'm sure it won't be long

On Jan.28 he said:
The novel coronavirus outbreak may reach its peak in the next 10 to 14 days,

and we didn't believe it. But the numbers now show it.

said on Sunday, amending his previous prediction on Jan 28 that the epidemic would
peak in seven or 10 days.

Prolonging the Spring Festival holidays, controlling traffic, temperature checks and education
as well as other government measures have effectively cut off the source of infection and
greatly reduced the spread of the virus,

------------------------------
so, why does he now say it peaks end Feb, when it had already peaked ?
As he had predicted.
It would go up for a second peak ?
Maybe due to people returning to work ? No, he said that would probably have little impact.
So what is it ?

And then it will peak and not go down after that until late April ???

I'm puzzled

==========================================


I found this :


Feb18, Human-to-human transmission has not been stopped in Wuhan: Zhong Nanshan
Zhong Nanshan on Tuesday warned the public to keep their waste pipes unobstructed as the novel coronavirus may spread through drainage systems.

Xinhua, Feb.12 , Zhong Nanshan said the epidemic peak should appear in mid-to-late February,
but the inflection point is still undetermined.

Zhong Nanshan on Feb.12
http://www.ecns.cn/news/society/2020...h6522839.shtml
they show the same picture as in the Feb.17 article , so did he say this
already on Feb.12 and just recalled it on Feb17 without looking at the new data
in all his traveling stress ?
Maybe just only adding the April-forecast after speaking with the modeler
[see below]

> The much-anticipated turning point of the novel coronavirus outbreak cannot be prediced

5 days later on Feb17 he could predict it

> There is a mathematics model, which is continuously being rectified according to the real
> situation, and climate factors are being considered.

aha, some secret Chinese math. modeling software program
which includes climate factors. Something that Western researchers claim
to be impossible

> Currently the new confirmed cases show a decreasing trend, and the mathematics
> model predicted that the turning point will arrive in mid- or late-February in southern China,

## at that time it had already peaked in Southern China.
## but only a few datapoints available, so with statistical uncertainety, "errorbar"
## they may have required the common 95% statistical errorbar

## and the April national prediction may be because it includes Wuhan,
## which "Southern China" does not

https://upload.wikimedia.org/wikiped...hina_map-1.png

Share on Facebook Share on Twitter