SERVING THE QUANTITATIVE FINANCE COMMUNITY

 
User avatar
Paul
Posts: 10787
Joined: July 20th, 2001, 3:28 pm

Re: How good or bad are the pandemic warning models?

April 25th, 2020, 4:22 pm

19th century here we come!
 
User avatar
Paul
Posts: 10787
Joined: July 20th, 2001, 3:28 pm

Re: How good or bad are the pandemic warning models?

April 25th, 2020, 5:29 pm

How do we distinguish causes of falling numbers?: lockdown, fewer susceptibles, weather, apathy,...
 
User avatar
Paul
Posts: 10787
Joined: July 20th, 2001, 3:28 pm

Re: How good or bad are the pandemic warning models?

April 25th, 2020, 6:08 pm

We saw it coming:
pandemic.png
As we did the Spanish flu:
influenza.png
 
User avatar
katastrofa
Posts: 9329
Joined: August 16th, 2007, 5:36 am
Location: Alpha Centauri

Re: How good or bad are the pandemic warning models?

April 25th, 2020, 7:53 pm

How do we distinguish causes of falling numbers?: lockdown, fewer susceptibles, weather, apathy,...
Isn't it enough to compare previous years' trends?
 
User avatar
Paul
Posts: 10787
Joined: July 20th, 2001, 3:28 pm

Re: How good or bad are the pandemic warning models?

April 25th, 2020, 8:21 pm

How do we distinguish causes of falling numbers?: lockdown, fewer susceptibles, weather, apathy,...
Isn't it enough to compare previous years' trends?
You mean for flu, say? 
 
User avatar
katastrofa
Posts: 9329
Joined: August 16th, 2007, 5:36 am
Location: Alpha Centauri

Re: How good or bad are the pandemic warning models?

April 25th, 2020, 8:38 pm

Yes. You just need to stratify the data by confounders.
 
User avatar
Paul
Posts: 10787
Joined: July 20th, 2001, 3:28 pm

Re: How good or bad are the pandemic warning models?

April 25th, 2020, 8:43 pm

That would be great, it would show lots of things, including how good the models are! But where do you get the data for lockdowns from?
 
User avatar
katastrofa
Posts: 9329
Joined: August 16th, 2007, 5:36 am
Location: Alpha Centauri

Re: How good or bad are the pandemic warning models?

April 25th, 2020, 9:40 pm

Naive solution:
1) Take the last decade's monthly mortality rate data on a set of relevant covariates (weather, GDP, unemployment rate, etc.) and stratify by age and other reported covid risk factors.
2) Run a linear regression for each group and extrapolate until today (or use my Deep Recurrent Neural Net model to forecast the trends :-D).
3) Take the reported death rates due to covid in the specified age groups and subtract from the extrapolation.
4) Compare the reported mortality rates due to all causes with your result to see the lockdown effect.
 
User avatar
katastrofa
Posts: 9329
Joined: August 16th, 2007, 5:36 am
Location: Alpha Centauri

Re: How good or bad are the pandemic warning models?

April 25th, 2020, 9:57 pm

We saw it coming:

pandemic.png

As we did the Spanish flu:

influenza.png
https://books.google.com/ngrams/graph?c ... wn%3B%2Cc0

:-O
The pen is mightier than the sword, they say!
BTW, Russian microbiologist claims Coronavirus is result of scientists doing 'crazy things' in Wuhan lab

"Nobel winning French professor Luc Montagnier recently claimed that the novel Coronavirus was created inside a lab and mentioned that the virus which infected over 2.7 million people globally was accidentally released from Wuhan Institution of Virology. Now, a leading Russian microbiologist also claimed that the novel Coronavirus is the result of Wuhan scientists doing "absolutely crazy things" in their laboratory.

The well-known Russian Professor Peter Chumakov said that the aim of Chinese scientists was to study the pathogenicity of the virus. But he clarified that it was not their intention to deliberately create a man-made killer."

How does he know their intentions??
 
User avatar
Paul
Posts: 10787
Joined: July 20th, 2001, 3:28 pm

Re: How good or bad are the pandemic warning models?

April 25th, 2020, 10:43 pm

Naive solution:
1) Take the last decade's monthly mortality rate data on a set of relevant covariates (weather, GDP, unemployment rate, etc.) and stratify by age and other reported covid risk factors.
2) Run a linear regression for each group and extrapolate until today (or use my Deep Recurrent Neural Net model to forecast the trends :-D).
3) Take the reported death rates due to covid in the specified age groups and subtract from the extrapolation.
4) Compare the reported mortality rates due to all causes with your result to see the lockdown effect.
I'd like to see all that! And, indeed, people are starting to do some plots like that. But if we have no clue how many susceptibles there are left...?

This plot is interesting
Figure 1_ The number of deaths involving COVID-19 and _Influenza and Pneumonia_ increased compared with the previous week.png
from https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/deathsregisteredweeklyinenglandandwalesprovisional/weekending3april2020 showing an extra 2,000 deaths not attributed to Covid-19. That's worrying! 
 
 
User avatar
Paul
Posts: 10787
Joined: July 20th, 2001, 3:28 pm

Re: How good or bad are the pandemic warning models?

April 25th, 2020, 10:51 pm

It reminds me of the assassination of Alexander Litvinenko. I found that the website (something like) polonium-210.com had been registered just one week before he was poisoned.
 
User avatar
katastrofa
Posts: 9329
Joined: August 16th, 2007, 5:36 am
Location: Alpha Centauri

Re: How good or bad are the pandemic warning models?

April 26th, 2020, 2:26 am

Actually, https://www.petoxygenmasks.org/4mypet.html (I ordered 3 in case the cats need it)
However since then I've learnt that it's not about breathing problems, but coronavirus destroys haem complexes: it knocks out the iron from the complex to use the remaining porphyrin ring to get to the cells (or just does it - which is ultimately its mistake...). Free haemoglobin can't bind oxygen to deliver it lungs. That's why patients have low oxygen levels in blood. The lungs are fine by then. The problem with lungs begins when the released iron catalyses reactions which produce free radicals. They attack lungs causing ARDS and damage other organs too.
That's why they should simultaneously give the covid patients blood transfusions and ground them. New haems will deliver oxygen, and free radicals will absorb free electrons flowing into the body from the Earth.
BTW, some medic on Medium "debunked" the mechanism of covid destroying heam claiming that there's no evidence and that he's a covid guru - I'm not sure what exactly he was debunking, but there are results of biopsies which suggest the patients developed anaemia and their spleens were depleted suggesting that the organ tried to compensate for the destroyed red blood cells, and also susceptibility to the virus depended on blood type... But I'm in the middle of my microbiological covid theory (the above is based on some less or more certain sources). TBC

Grounding is such an old medical practice, but nobody does it in the West: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4378297/
I have natural instinct to lie down on the ground naked when I feel ill, back at home I practised it regularly (in the forest where nobody could see me).
 
User avatar
Cuchulainn
Posts: 62395
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Re: How good or bad are the pandemic warning models?

April 28th, 2020, 11:44 am

'Worrying' number of Covid-19 cases in border counties - Dr Gabriel Scally

https://www.rte.ie/news/coronavirus/2020/0428/1135264-covid19-coronavirus-border/
 
User avatar
Cuchulainn
Posts: 62395
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Re: How good or bad are the pandemic warning models?

April 28th, 2020, 11:47 am

Actually, https://www.petoxygenmasks.org/4mypet.html (I ordered 3 in case the cats need it)
However since then I've learnt that it's not about breathing problems, but coronavirus destroys haem complexes: it knocks out the iron from the complex to use the remaining porphyrin ring to get to the cells (or just does it - which is ultimately its mistake...). Free haemoglobin can't bind oxygen to deliver it lungs. That's why patients have low oxygen levels in blood. The lungs are fine by then. The problem with lungs begins when the released iron catalyses reactions which produce free radicals. They attack lungs causing ARDS and damage other organs too.
That's why they should simultaneously give the covid patients blood transfusions and ground them. New haems will deliver oxygen, and free radicals will absorb free electrons flowing into the body from the Earth.
BTW, some medic on Medium "debunked" the mechanism of covid destroying heam claiming that there's no evidence and that he's a covid guru - I'm not sure what exactly he was debunking, but there are results of biopsies which suggest the patients developed anaemia and their spleens were depleted suggesting that the organ tried to compensate for the destroyed red blood cells, and also susceptibility to the virus depended on blood type... But I'm in the middle of my microbiological covid theory (the above is based on some less or more certain sources). TBC

Grounding is such an old medical practice, but nobody does it in the West: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4378297/
I have natural instinct to lie down on the ground naked when I feel ill, back at home I practised it regularly (in the forest where nobody could see me).
Dangerous; nettles and ticks.
ABOUT WILMOTT

PW by JB

Wilmott.com has been "Serving the Quantitative Finance Community" since 2001. Continued...


Twitter LinkedIn Instagram

JOBS BOARD

JOBS BOARD

Looking for a quant job, risk, algo trading,...? Browse jobs here...


GZIP: On