Coronavirus - Data Analysis Thread

Coronavirus - Data Analysis Thread

Author
Discussion

Elysium

Original Poster:

10,714 posts

154 months

Monday 21st December 2020
quotequote all
bigandclever said:
coanda said:
bigandclever said:
That’s not data, it’s a terrible, unintelligible chart.
Let's see your attempt at answering Otis' question then.
It’s unreadable. Maybe it was great on the poster’s screen, but by the time it gets to mine it’s illegible.

eg

You have clicked on it once to open it in thumbsnap. Click on it again to make it high res. It’s how the image hosting here works unfortunately. The link to the website that produced it is earlier in the thread.

bigandclever

11,460 posts

205 months

Monday 21st December 2020
quotequote all
Elysium said:
You have clicked on it once to open it in thumbsnap. Click on it again to make it high res. It’s how the image hosting here works unfortunately. The link to the website that produced it is earlier in the thread.
Bloody hell, ok, user error smile

Elysium

Original Poster:

10,714 posts

154 months

Monday 21st December 2020
quotequote all
BlackTails said:
smartypants said:
Just data. No hypothesis.
Good idea. And yet the very next post contained the following hypotheses:

Elysium said:
This shows the vast scale of infections that was likely in the spring, but that we were completely unable to detect. It also shows that, despite massive increases in testing and case numbers we must still be missing a great many cases in the autumn.

Admissions and are deaths barely visible at this scale.
A data only thread is a good idea but unattainable. The temptation to provide a commentary (opinion, speculation or conclusion, take your pick) on the data is overwhelming and irresistible.

So, a suggestion: any interpretation - any - is prefaced by “in my opinion”, or “in my view”, or “I infer” or something similar. If you’re not sure whether what you’re saying is fact or your view of the facts, err on the side of caution.

Counting down 3-2-1 for this idea to be roundly ignored...
In my first post and in the title of the thread I suggested this should be for the analysis of the data. That does require some commentary to be made,

I took particular care to explain how this curve had been calculated, because it is imputed, not measured data. However, I have clarified my comment as per your suggestion.


Thin White Duke

2,002 posts

127 months

Monday 21st December 2020
quotequote all
The graphs made and posted by some of the members, in particular Elysium have been great. However all the information used is readily available to the public, even the BBC have produced graphs which show the "real picture."

It just seems that Joe public doesn't much care to look into these things for themselves and only take in the doom mongering graphs presented by Whitty and Valance.

Anyway keep up the good work guys.

Tuna

18,686 posts

251 months

LimJim

2,274 posts

9 months

Monday 21st December 2020
quotequote all

Otispunkmeyer

11,218 posts

122 months

Tuesday 22nd December 2020
quotequote all
Elysium said:
Thanks. I think that answers the question. There was a dip right after (or they were already in a downward trajectory) but it has risen again in the proceeding week. What is the general idea? 7 days to full blown symptoms and about 14 for the deaths?

Perhaps some was pre-baked, but I think I would err on the side of those anecdotes not being correct.

Elysium

Original Poster:

10,714 posts

154 months

Tuesday 22nd December 2020
quotequote all
Country comparisons are difficult, but I think we need to do more to understand why east asian countries are experiencing a very different pandemic to Europe and the USA:


Elysium

Original Poster:

10,714 posts

154 months

Tuesday 22nd December 2020
quotequote all
When comparing countries Sweden and UK normally come up because the timing of their epidemic curves is similar and they have different lockdown policies. People also argue that Sweden should be compared to neighbouring states.

This is from Euromomo, where you can choose countries to compare z-scores and zoom in on a particular timescale. In this case all of 2020:

https://www.euromomo.eu/graphs-and-maps/#



UK lockdown was wk 13, so if we assume 3 weeks from infection to death the impacts should be seen wk 16.

This comparison between Sweden and neighbouring states suggests they faced difficult levels of infection between wks 13 and 16.

Elysium

Original Poster:

10,714 posts

154 months

Tuesday 22nd December 2020
quotequote all
ONS data published today:

https://www.ons.gov.uk/peoplepopulationandcommunit...

This shows ONS all cause deaths with those 'due to' COVID highlighted in orange. The dotted lines are the average, max and min for 2015-2019 adjusted for the 2019 population. The solid green line is the limit of statistical significance or 2 x standard deviations from the average.

I think it is interesting that we had sigificant non-COVID excess death in the spring, but non-COVID deaths are now below the 5 year minimum. This requires further explanation.



This second graph uses the same data set, to show 'due to COVID' deaths and excess deaths. I think this gives a good idea of the relative scale of the spring and autumn outbreaks:


survivalist

3,741 posts

157 months

Tuesday 22nd December 2020
quotequote all
Otispunkmeyer said:
Elysium said:
This graph shows the 7 day average of cases for London and Wales. I have identified the time periods where the Welsh Firebreak and English Lockdown could have had an impact on transmission.

Data here: https://coronavirus.data.gov.uk/details/cases

Both show a dip during tighter restrictions with a rapid rebound. London has rebounded harder, but was growing faster prior to restrictions. The dip at the end is reporting lag:



Edited by Elysium on Sunday 20th December 15:28
Like I replied in the other thread. Looks like lockdown does have an effect. It is just short lived and doesn't effect the eventual outcome given what looks like a quick "reversion to the mean" once the restrictions stop.

So perhaps there is a use for them. But the expectations should be that it's not going to make the virus go away. It will come straight back again the moment you emerge. Obviously we cannot lockdown ad infinitum, but perhaps they can be used to buy some time for the services.

But then we are not seeing anything like as bad as it was in spring, so are they actually being genuinely useful?
Do they publish the number of tests that are carried out on a daily basis by region? If so it would be interested if this was mapped against this graph. Just wondering if, when locked down, fewer people would have a test? After all, if you're at home with no plans to go out for a couple of weeks, what's the point in getting tested anyway?

ETA - or more specifically, the % of positive tests results. Would be interested to see how they are linked.


Edited by survivalist on Tuesday 22 December 12:20

Elysium

Original Poster:

10,714 posts

154 months

Tuesday 22nd December 2020
quotequote all
survivalist said:
Otispunkmeyer said:
Elysium said:
This graph shows the 7 day average of cases for London and Wales. I have identified the time periods where the Welsh Firebreak and English Lockdown could have had an impact on transmission.

Data here: https://coronavirus.data.gov.uk/details/cases

Both show a dip during tighter restrictions with a rapid rebound. London has rebounded harder, but was growing faster prior to restrictions. The dip at the end is reporting lag:



Edited by Elysium on Sunday 20th December 15:28
Like I replied in the other thread. Looks like lockdown does have an effect. It is just short lived and doesn't effect the eventual outcome given what looks like a quick "reversion to the mean" once the restrictions stop.

So perhaps there is a use for them. But the expectations should be that it's not going to make the virus go away. It will come straight back again the moment you emerge. Obviously we cannot lockdown ad infinitum, but perhaps they can be used to buy some time for the services.

But then we are not seeing anything like as bad as it was in spring, so are they actually being genuinely useful?
Do they publish the number of tests that are carried out on a daily basis by region? If so it would be interested if this was mapped against this graph. Just wondering if, when locked down, fewer people would have a test? After all, if you're at home with no plans to go out for a couple of weeks, what's the point in getting tested anyway?
This shows a couple of dips in testing, but only the second of these coincides with the period where the second English lockdown would have an impact on cases:


survivalist

3,741 posts

157 months

Tuesday 22nd December 2020
quotequote all
Elysium said:
survivalist said:
Otispunkmeyer said:
Elysium said:
This graph shows the 7 day average of cases for London and Wales. I have identified the time periods where the Welsh Firebreak and English Lockdown could have had an impact on transmission.

Data here: https://coronavirus.data.gov.uk/details/cases

Both show a dip during tighter restrictions with a rapid rebound. London has rebounded harder, but was growing faster prior to restrictions. The dip at the end is reporting lag:



Edited by Elysium on Sunday 20th December 15:28
Like I replied in the other thread. Looks like lockdown does have an effect. It is just short lived and doesn't effect the eventual outcome given what looks like a quick "reversion to the mean" once the restrictions stop.

So perhaps there is a use for them. But the expectations should be that it's not going to make the virus go away. It will come straight back again the moment you emerge. Obviously we cannot lockdown ad infinitum, but perhaps they can be used to buy some time for the services.

But then we are not seeing anything like as bad as it was in spring, so are they actually being genuinely useful?
Do they publish the number of tests that are carried out on a daily basis by region? If so it would be interested if this was mapped against this graph. Just wondering if, when locked down, fewer people would have a test? After all, if you're at home with no plans to go out for a couple of weeks, what's the point in getting tested anyway?
This shows a couple of dips in testing, but only the second of these coincides with the period where the second English lockdown would have an impact on cases:

Interesting, thanks.

The next couple of weeks will be interesting. Assuming test capacity remains broadly flat, I’m curious to see if actual tests drop over the Christmas & New Year period as people focus on fun.

Guess the schools and universities being closed will also have an impact.

Elysium

Original Poster:

10,714 posts

154 months

Tuesday 22nd December 2020
quotequote all
It’s effectively the ONS excess deaths data, but I think this is a great ‘Florence Nightingale’ graph shared by Carl Heneghan:



https://www.cebm.net/covid-19/covid-19-florence-ni...

LimJim

2,274 posts

9 months

Wednesday 23rd December 2020
quotequote all

Isaac Hunt

12,900 posts

178 months

Wednesday 23rd December 2020
quotequote all
Stolen from Grumbleoak in the other thread



https://twitter.com/simongerman600/status/13414132...

WindyCommon

2,586 posts

206 months

Wednesday 23rd December 2020
quotequote all
Thread is holding up well..!

Elysium

Original Poster:

10,714 posts

154 months

Wednesday 23rd December 2020
quotequote all
So far yes

Just seen that the Gov staging site now provides positivity data at local authority level for England:

https://coronavirus.data.gov.uk/details/testing

Hastings, which is caught up in the 'Southern Surge' is hitting 15%



However, Liverpool and Nottingham were 19.9% and 32.5% respectively when they peaked:





Interesting that positivity remained steady during Liverpools mass testing spike.

Otispunkmeyer

11,218 posts

122 months

Wednesday 23rd December 2020
quotequote all
Elysium said:
Country comparisons are difficult, but I think we need to do more to understand why east asian countries are experiencing a very different pandemic to Europe and the USA:

Unless everyone records data in the same way it's probably going to be difficult to tell.

Bodo

11,202 posts

233 months

Wednesday 23rd December 2020
quotequote all
LimJim said:
Sharing Chinese state media information in a Western forum is like Kim Jong un speaking to his people; only in English.

Please be careful with possibly manipulative information sources.