Ski Club 2.0 Home
Snow Reports
FAQFAQ

Mail for help.Help!!

Log in to snowHeads to make it MUCH better! Registration's totally free, of course, and makes snowHeads easier to use and to understand, gives better searching, filtering etc. as well as access to 'members only' forums, discounts and deals that U don't even know exist as a 'guest' user. (btw. 50,000+ snowHeads already know all this, making snowHeads the biggest, most active community of snow-heads in the UK, so you'll be in good company)..... When you register, you get our free weekly(-ish) snow report by email. It's rather good and not made up by tourist offices (or people that love the tourist office and want to marry it either)... We don't share your email address with anyone and we never send out any of those cheesy 'message from our partners' emails either. Anyway, snowHeads really is MUCH better when you're logged in - not least because you get to post your own messages complaining about things that annoy you like perhaps this banner which, incidentally, disappears when you log in :-)
Username:-
 Password:
Remember me:
👁 durr, I forgot...
Or: Register
(to be a proper snow-head, all official-like!)

Is the 2020/2021 a non starter?

 Poster: A snowHead
Poster: A snowHead
No-one has yet to discuss our basic freedoms either. It has been pointed out in the Press that for years pubs shut at 11 o'clock because of an old WW1 rule no-one had the wit to bin. It took until Thatcher to aggressively prune back the State. Expect more interference in our personal freedom to choose and act, all in the name of it being 'for your own good'. Never mind the sugar tax etc this is going to be full on. It will only be resisted if the more bovine elements of the Public push back which given our most recent performance I strongly doubt. There will be plastic screens on shop tills for years yet.

"Of all tyrannies, a tyranny sincerely exercised for the good of its victims may be the most oppressive. It would be better to live under robber barons than under omnipotent moral busybodies. The robber baron's cruelty may sometimes sleep, his cupidity may at some point be satiated; but those who torment us for our own good will torment us without end for they do so with the approval of their own conscience. They may be more likely to go to Heaven yet at the same time likelier to make a Hell of earth. This very kindness stings with intolerable insult. To be "cured" against one's will and cured of states which we may not regard as disease is to be put on a level of those who have not yet reached the age of reason or those who never will; to be classed with infants, imbeciles, and domestic animals."

C.S. Lewis
latest report
 Obviously A snowHead isn't a real person
Obviously A snowHead isn't a real person
justabod wrote:
@peanuthead, keep staff groups separate. Death is far more certain from many cancers than from C19. Many elderly stay in a lot of the time anyway, and why should they not be entitled to take the risk? They have little life left and take a risk when they go out every 'flu season yet they still do it. 14/15 'flu season killed 24k and that was WITH a vaccine. We don't turn off the economy each winter.

Besides, how do you know the hospitals would be overwhelmed? Large quantities of deaths are happening in care and nursing homes, most of those I see there already have DNARs and advanced directives which specifically state "not for hospital". They were going to die in the NH anyway. Again, I refer to the basic concept of triage: ITU overwhelmed means a smaller number die than those who would from elective and even emergency surgery. This is type II thinking that is necessary to run a health system or country. I'm not some great strategist but I have at least the ability to look at it coldly.


Agree, too many are running round like scared cats.
If you are under 65, and or arent obese, dark skinned, have sickle cell, heart issues, lung problems or diabetes the chances of needing hospital treatment and very small and dying are tiny, compared to death rates from smoking etc it's not even a blip on the stats.
Personally I blame the media for a lot of this panic
latest report
 Well, the person's real but it's just a made up name, see?
Well, the person's real but it's just a made up name, see?
So the elderly are completely expendable and don't matter at all because they have had their lives?

Wow... Sad

Never mind applauding the NHS, you'll be out applauding Boris's handling of corona virus in care homes next...
snow report
 You need to Login to know who's really who.
You need to Login to know who's really who.
mountainaddict wrote:


Never mind applauding the NHS, you'll be out applauding Boris's handling of corona virus in care homes next...


It's just a question of an Orwellian press control. There's going to be an awful lot of "double-think" at some point. However the Euro-view is highly critical of this government's slow response to an obviously pressing pandemic ….... https://www.euronews.com/2020/05/06/coronavirus-why-is-the-uk-s-covid-19-death-toll-higher-than-other-eu-countries

The spread of Covid-19 in the UK is proportional to the delay in locking down.

Compare now the UK deaths with Austria, Germany, and Switzerland in terms of deaths per 100,000 population.

More worrying is the math regarding the number of UK positive test results, and the number of deaths of those testing positive. Currently stands at just over 14%!
latest report
 Anyway, snowHeads is much more fun if you do.
Anyway, snowHeads is much more fun if you do.
@endoman, I really like your post above.

I'll add one thing. Too often people use the appearance of system 2 thinking to justify their system 1 already-arrived at conclusions, whichever they are. The easiest way to spot this is that people don't change their opinion when the data changes (and few admit they may have been wrong).

This is happening on both sides of the debate, including here on snowheads. It's quite interesting (and disheartening) to see.
snow report
 You'll need to Register first of course.
You'll need to Register first of course.
skimastaaah wrote:
mountainaddict wrote:


Never mind applauding the NHS, you'll be out applauding Boris's handling of corona virus in care homes next...


It's just a question of an Orwellian press control. There's going to be an awful lot of "double-think" at some point. However the Euro-view is highly critical of this government's slow response to an obviously pressing pandemic ….... https://www.euronews.com/2020/05/06/coronavirus-why-is-the-uk-s-covid-19-death-toll-higher-than-other-eu-countries

The spread of Covid-19 in the UK is proportional to the delay in locking down.

Compare now the UK deaths with Austria, Germany, and Switzerland in terms of deaths per 100,000 population.

More worrying is the math regarding the number of UK positive test results, and the number of deaths of those testing positive. Currently stands at just over 14%!


The more tests you do the more positive tests you will get. Thought UK policy was only to test people with symptoms at present as well as NHS workers? Hence of course the positive test as a proportion of total tests will be higher due to the sample tested. Start testing lots of symptom free people and that percentage will drop. But at 14% of the approx 100K per day it means the disease is rife in the population now. At 100,000 tests a day it will take almost 2 years to test the population.....
snow report
 Then you can post your own questions or snow reports...
Then you can post your own questions or snow reports...
Quote:

The spread of Covid-19 in the UK is proportional to the delay in locking down.

Compare now the UK deaths with Austria, Germany,.....

Worryingly, Beeb lunchtime news reported that there's been a spike in German cases since they've been relaxing their lockdown.

Have to see whether Dr John Campbell reports on this on today's world report on YouTube
latest report
 After all it is free Go on u know u want to!
After all it is free Go on u know u want to!
skimastaaah wrote:
The spread of Covid-19 in the UK is proportional to the delay in locking down.

Compare now the UK deaths with Austria, Germany, and Switzerland in terms of deaths per 100,000 population.
I think the UK and Germany locked down at about the same time (one day apart?). The picture in Germany slightly complicated by different States having slightly different responses, but broadly speaking no difference between when the UK and Germany went in to lockdown as I understand it.
snow report
 You'll get to see more forums and be part of the best ski club on the net.
You'll get to see more forums and be part of the best ski club on the net.
@mountainaddict, no. I've neither suggested nor asserted that. If you are 85 you have had your life, at least the greater portion of it. You have had your chances to be educated, to choose a profession, to marry and have children then rear them. Whatever has been done with that time, it was there. A child or even middle-aged man has not had their chances and have yet to grow old. Everyone has their lot of time and we all aim for as much as possible, expecting to live until our eighties or more nowadays.

I go into care and nursing homes on an almost daily basis. They have been handled poorly but there are thousands of them and are private businesses. Some have been wise, others not. There is no overarching organisation. I would suggest that many are better and more humanely run than hospitals. Many care homes close or are significantly changed after failure (Winterbourne View). Mid Staffs is still open and mostly unchanged (I'll bet) since the appalling neglect that was exposed.

@skimastaaah, incorrect. As I've mentioned before some countries have had a very light quarantine: see SK, Taiwan and Sweden. They should all be dead surely? Italy and Spain had v early quarantines yet many dead.

@horizon, fair point. It's worth pointing out however that the data is very wobbly. Ferguson's code has been torn apart yet this was used by the gov't to justify the ruining of our freedoms. If the gov't believed the figures they would be reminding us of the dead if we'd done nothing (550k). Even with quarantine the numbers were supposed to be much higher, does anyone remember what they were supposed to be? At least hundreds of thousands I think.
snow report
 Ski the Net with snowHeads
Ski the Net with snowHeads
horizon wrote:
@endoman, I really like your post above.

I'll add one thing. Too often people use the appearance of system 2 thinking to justify their system 1 already-arrived at conclusions, whichever they are. The easiest way to spot this is that people don't change their opinion when the data changes (and few admit they may have been wrong).

This is happening on both sides of the debate, including here on snowheads. It's quite interesting (and disheartening) to see.


Changing your views depending on the available data is absolutely essential. Sadly in this situation, some people's salaries depend on them not changing their opinions. Ferguson being a classic.
ski holidays
 snowHeads are a friendly bunch.
snowHeads are a friendly bunch.
There is a fantastic set of extended interviews with experts on the Unherd website, including Prof Ferguson and other virologists who are highly critical of him. Should be required viewing before making online comments!!
The view is heavily based towards lockdowns just spreading out the same number of deaths over a longer period of time-given true lack of vaccine for the foreseeable future and the social/economic unsustainability of lockdowns. Lockdowns don’t change the biology or either us or the virus.
latest report
 And love to help out and answer questions and of course, read each other's snow reports.
And love to help out and answer questions and of course, read each other's snow reports.
One by Prof Levitt also goes into the uncomfortable but necessary distinctions between the death of an 85 year old who is ill and likely to die soon anyway, and that of a 25 year old with their life ahead of them.


Last edited by And love to help out and answer questions and of course, read each other's snow reports. on Sun 10-05-20 16:29; edited 1 time in total
snow report
 So if you're just off somewhere snowy come back and post a snow report of your own and we'll all love you very much
So if you're just off somewhere snowy come back and post a snow report of your own and we'll all love you very much
skimastaaah wrote:
mountainaddict wrote:


Never mind applauding the NHS, you'll be out applauding Boris's handling of corona virus in care homes next...


It's just a question of an Orwellian press control. There's going to be an awful lot of "double-think" at some point. However the Euro-view is highly critical of this government's slow response to an obviously pressing pandemic ….... https://www.euronews.com/2020/05/06/coronavirus-why-is-the-uk-s-covid-19-death-toll-higher-than-other-eu-countries

The spread of Covid-19 in the UK is proportional to the delay in locking down.

Compare now the UK deaths with Austria, Germany, and Switzerland in terms of deaths per 100,000 population.

More worrying is the math regarding the number of UK positive test results, and the number of deaths of those testing positive. Currently stands at just over 14%!


What delay,? Every country in europe that has lockdowned has done do between 51 and 54 days after first confirmed case, how about comparing deaths per 1000 with belgium, Spain or italy.
And no one knows who will end up doing best until we can compare excess deaths this year compared to the average, then of course there are other reasons why some are more likely to succumb, such as genetics, general health etc
ski holidays
 You know it makes sense.
You know it makes sense.
@endoman, Ferguson wasn't paid for his role as a government advisor. Perhaps check the facts before posting bolloxs next time
snow conditions
 Otherwise you'll just go on seeing the one name:
Otherwise you'll just go on seeing the one name:
rambotion wrote:
@endoman, Ferguson wasn't paid for his role as a government advisor. Perhaps check the facts before posting bolloxs next time


He was paid for his work at Imperial, and was out by orders of magnitude, more than once.

https://www.aier.org/article/how-wrong-were-the-models-and-why/
ski holidays
 Poster: A snowHead
Poster: A snowHead
snowhound wrote:
The view is heavily based towards lockdowns just spreading out the same number of deaths over a longer period of time-given true lack of vaccine for the foreseeable future and the social/economic unsustainability of lockdowns. Lockdowns don’t change the biology or either us or the virus.


That's a curious conclusion to come to. If Lockdowns merely spread the same number of deaths over a longer period of time, then why both with them at all? Surely one should not lockdown at all, have the total deaths within a 3week period, and then become the new super-power of the world by the end of the month?
Ah no... because the point of lockdowns is to keep infections at a low enough level so that those people that COULD survive are able to get into a hospital and get treatment. If you can treat and save only 5% of those that get infected and need hospitalisation, then your lockdown efforts have prevented deaths.
ski holidays
 Obviously A snowHead isn't a real person
Obviously A snowHead isn't a real person
Quote:
That's a curious conclusion to come to. If Lockdowns merely spread the same number of deaths over a longer period of time, then why both with them at all? .

That's the whole point of lockdown, to spread the load so the NHS isn't overwhelmed - same number of cases (area under graph) but over a longer time

snow report
 Well, the person's real but it's just a made up name, see?
Well, the person's real but it's just a made up name, see?
Sorry I should have confirmed that the one condition to maintain was to ensure the health service could manage the numbers to ensure extra people did not die because they were overwhelmed. With that condition, the number of deaths would be the same. But they were pretty much categoric that eradication/elimination/complete suppression was not possible in an open society with an endemic virus.
But even if more people did die you need balance with the social and economic damage, and all the deaths caused by other illnesses. What odds on a wave of cancer diagnosis and deaths in 18-24 months time?


Last edited by Well, the person's real but it's just a made up name, see? on Sun 10-05-20 18:56; edited 1 time in total
snow conditions
 You need to Login to know who's really who.
You need to Login to know who's really who.
endoman wrote:
rambotion wrote:
@endoman, Ferguson wasn't paid for his role as a government advisor. Perhaps check the facts before posting bolloxs next time


He was paid for his work at Imperial, and was out by orders of magnitude, more than once.

https://www.aier.org/article/how-wrong-were-the-models-and-why/


He's an academic. It is irrelevant to his salary if he is right or wrong. Don't forget that Ferguson et al only provide advice to government, and that decisions (and responsibility) rest with them.
snow report
 Anyway, snowHeads is much more fun if you do.
Anyway, snowHeads is much more fun if you do.
In which case advice should change when models are proven to be erroneous.
latest report
 You'll need to Register first of course.
You'll need to Register first of course.
endoman wrote:
In which case advice should change when models are proven to be erroneous.
Is there academic acceptance that the Imperial College team's models are erroneous?
snow conditions
 Then you can post your own questions or snow reports...
Then you can post your own questions or snow reports...
As a scientist myself, the main scandal is that his model, its code, its results and the data/assumptions used have never been peer assessed and put under full evaluation. Despite this it has had a huge impact on social and economic policy. A black box exercise
In decision making. Whether his model is correct or not, that lack of scientific and political process-not exercising due caution-is completely wrong.
snow conditions
 After all it is free Go on u know u want to!
After all it is free Go on u know u want to!
snowhound wrote:
As a scientist myself, the main scandal is that his model, its code, its results and the data/assumptions used have never been peer assessed and put under full evaluation. Despite this it has had a huge impact on social and economic policy. A black box exercise
In decision making. Whether his model is correct or not, that lack of scientific and political process-not exercising due caution-is completely wrong.
Fully agree.
ski holidays
 You'll get to see more forums and be part of the best ski club on the net.
You'll get to see more forums and be part of the best ski club on the net.
There is numerical evidence. I’m in system 2 now. The best case predicted in their paper was 1.3 million deaths worldwide, worst case 40 million. At end of March when published, death total was 102k. It’s now 281k and we are well into the tail of the distribution. We still need 1million more deaths to meet best case Scenario. The paper is here.

https://www.imperial.ac.uk/mrc-global-infectious-disease-analysis/covid-19/report-12-global-impact-covid-19/

I suspect many academics are questioning this. I’m not an academic, but have been fortunate enough to listen to many world class statisticians, epidemiologists and cig eco people for the last ten years.
snow report
 Ski the Net with snowHeads
Ski the Net with snowHeads
snowhound wrote:
As a scientist myself, the main scandal is that his model, its code, its results and the data/assumptions used have never been peer assessed and put under full evaluation. Despite this it has had a huge impact on social and economic policy. A black box exercise
In decision making. Whether his model is correct or not, that lack of scientific and political process-not exercising due caution-is completely wrong.


100% agree
snow report
 snowHeads are a friendly bunch.
snowHeads are a friendly bunch.
That said I have seen quite a lot of criticism of the way in which the Imperial model was constructed from professional ( as opposed to academic) mathematical modellers.
The code they did release last week under pressure was materially different to that used in the March model, which does not inspire confidence.
snow report
 And love to help out and answer questions and of course, read each other's snow reports.
And love to help out and answer questions and of course, read each other's snow reports.
@snowhound, agreed. Unherd is a superb website although I find their "five min read" thing annoying: for those with limited attention spans, like small children. They have some excellent writers. I was initially drawn there by Douglas Murray.
snow report
 So if you're just off somewhere snowy come back and post a snow report of your own and we'll all love you very much
So if you're just off somewhere snowy come back and post a snow report of your own and we'll all love you very much
rob@rar wrote:
skimastaaah wrote:
The spread of Covid-19 in the UK is proportional to the delay in locking down.

Compare now the UK deaths with Austria, Germany, and Switzerland in terms of deaths per 100,000 population.
I think the UK and Germany locked down at about the same time (one day apart?). The picture in Germany slightly complicated by different States having slightly different responses, but broadly speaking no difference between when the UK and Germany went in to lockdown as I understand it.


So that's why UK deaths are 31,800 and Germany 7,500.

The delay included 3 days at Cheltenham, and the Liverpool/Madrid footie match. Most certainly a 6 day delay!

Currently the UK has the second highest deaths globally. Whichever way you want to defend this government's strategies, the bare facts are chilling.
latest report
 You know it makes sense.
You know it makes sense.
skimastaaah wrote:
rob@rar wrote:
skimastaaah wrote:
The spread of Covid-19 in the UK is proportional to the delay in locking down.

Compare now the UK deaths with Austria, Germany, and Switzerland in terms of deaths per 100,000 population.
I think the UK and Germany locked down at about the same time (one day apart?). The picture in Germany slightly complicated by different States having slightly different responses, but broadly speaking no difference between when the UK and Germany went in to lockdown as I understand it.


So that's why UK deaths are 31,800 and Germany 7,500.

The delay included 3 days at Cheltenham, and the Liverpool/Madrid footie match. Most certainly a 6 day delay!

Currently the UK has the second highest deaths globally. Whichever way you want to defend this government's strategies, the bare facts are chilling.


Depends how deaths are accounted for, we will only know in 12 to 18 months times when we can count excess deaths.
Whether it's true or not I haven't checked but I saw one stat that showed while deaths in the UK from pneumonia had plummeted in Germany they were rocketing. Too early to be definite
snow report
 Otherwise you'll just go on seeing the one name:
Otherwise you'll just go on seeing the one name:
@robs1, …. Mortality for COVID-19 appears higher than for influenza, especially seasonal influenza. While the true mortality of COVID-19 will take some time to fully understand, the data we have so far indicate that the crude mortality ratio (the number of reported deaths divided by the reported cases) is between 3-4%, the infection mortality rate (the number of reported deaths divided by the number of infections) will be lower. For seasonal influenza, mortality is usually well below 0.1%. However, mortality is to a large extent determined by access to and quality of health care.
ski holidays
 Poster: A snowHead
Poster: A snowHead
rob@rar wrote:
endoman wrote:
In which case advice should change when models are proven to be erroneous.
Is there academic acceptance that the Imperial College team's models are erroneous?


There is the model and there is the implementation of the model. The model itself cannot be "run" (in the sense of making widescale predictions) without the aid of a computer but obviously the maths can be proven although I'm not convinced the ICL team have the skills to do this. Once the math is translated into an algorithm there are tools to prove correctness although it is a long time since I've worked in this field so am not up to date.

It is clear the implementation is buggy.

For example in multi-threaded mode it produces random results. There is a difference between randomizing inputs and producing random results which you cannot explain. Epidemiologists who have commented on this point don't seem to understand the difference between the two which gives little confidence in what is referred to in the industry as "academic coding".

Academically speaking though. The original code has not been released. It was 15K lines of undocumented 'C' in a single file. There were no regression or unit tests, or rather ICL tried to write some automated tests but they never worked. It was coded by Fergussion at night after a hard day at the office. All of these are serious red flags and no-one should rely on this code for anything. That doesn't mean it doesn't run the model correctly in single user mode but you'd not want to fly an aircraft running this code, or base any decisions on it. In general code should be peer reviewed just as an academic article would be peer reviewed. That fact that the ICL team have skipped this important step shows how little they understand the process.

Another issue was back in March Fergusson was feeding garbage data into his model. He didn't know values like RO etc. GIGO even with the best model and implementation.

The ICL model has been demonstrably wrong for the other outbreaks where it has been used. Swine flu, Mad Cow, or at least Fergusson's interpretation of results have been wrong - widely alarmist.

If lessons are to be learned then there needs to be much stricter oversight of academics in future, at least for anything mission critical. A lot of countries based their decisions on the ICL results.
snow report
 Obviously A snowHead isn't a real person
Obviously A snowHead isn't a real person
@davidof, I am not sure whether "erroneous" is definable in this case. There was a widely quoted comment a couple of months ago, "all models are wrong, but some are useful". The question is not whether Ferguson's model is wrong - which it is bound to be - but whether it is useful.

The whole point of a model is to be able to input various hypothetical circumstances and see how they change what might happen. Right now that would be incredibly useful for working out what social distancing restrictions can be relaxed without risking a large change in R. However quite apart from the imperfection of any computer model itself, the circumstances input to the model will never be reflected in real life, particularly since humans don't behave in quite the way politicians tell them to. A model is not a prediction of what will happen, it is a way of identifying the range of what might happen.

However I do agree with you that it is extremely poor for any academic science not to include sufficient technical details for others to be able to replicate it. These days that can easily include providing access to the full model as an online technical appendix to a publication, but perhaps more important to provide an outline of the computational processes involved.
latest report
 Well, the person's real but it's just a made up name, see?
Well, the person's real but it's just a made up name, see?
Quote:

It’s now 281k and we are well into the tail of the distribution.

We are into the tail of this current initial phase. But infection rates are now headed up again in Germany, China and South Korea - to name just three of those countries thought by some to have sorted it out.

And Covid is only slowly makings its presence felt in many countries, including the Pacific Islands. Kiribati, for example, which is one of the most densely populated countries in the world, and has very high rates of diabetes and TB.

I agree with the arguments that the ICL models should be subject to peer review etc but the suggestion that his "best case" death predictions are wildly wrong is wildly premature.
latest report
 You need to Login to know who's really who.
You need to Login to know who's really who.
j b wrote:
@davidof, I am not sure whether "erroneous" is definable in this case. There was a widely quoted comment a couple of months ago, "all models are wrong, but some are useful". The question is not whether Ferguson's model is wrong - which it is bound to be - but whether it is useful.



I think you are falling into the same trap as the Epidemiologists - the numbers it spits out look fairly convincing based on sticking your finger in the air but if the implementation of the model is significantly buggy you can't base anything on the figures it produces to make decisions (Note: we don't know if it is buggy in single thread mode, we don't have the code, ICL says it doesn't work in multithread mode).

What would "erroneous" mean in terms of the model? At a basic level you define a function to model transmission rates base on RO but your function is mathematically incorrect.
You then have the implementation of that function in code, is that correct? Do you have unit tests to make sure that works and if updates are made to the code that code is not broken? Do you have functional tests to show how that works in unison with other parts of the code? It seems like ICL zapped a lot of that.
Epidemiologists tell everyone "to keep in their lanes" but then, like sorcerers apprentices they are messing around in other people's lanes.

Rob asked the question whether the academics had criticised the ICL model and I commented that you cannot separate the model from its implementation - the thing that gives you some data - in the way it is used. Academically the implementation hasn't been peer reviewed and what little we know (ICL are keeping the March code secret) it is of unacceptable quality to be trusted.

The model, or at least Fergusson's interpretation of his model, have been used to make some bad decisions in the past. So I would argue that it also needs looking closely based on empirical evidence.

The whole team sounds out of control and very reluctant to have any oversight. In industry this would be a huge red flag for me.


Last edited by You need to Login to know who's really who. on Mon 11-05-20 10:06; edited 2 times in total
latest report
 Anyway, snowHeads is much more fun if you do.
Anyway, snowHeads is much more fun if you do.
@davidof, thanks. I'm not anywhere near qualified to comment on the technicalities you raise, but certainly agree with your point that these things should be open to scrutiny.
ski holidays
 You'll need to Register first of course.
You'll need to Register first of course.
rob@rar wrote:
@davidof, thanks. I'm not anywhere near qualified to comment on the technicalities you raise, but certainly agree with your point that these things should be open to scrutiny.


but the problem is political, Boris chose to go with Fergusson's predictions, that was his choice.

The code (at least in single thread mode) may produce the results that the model says it should. The predictions sound plausible.
snow report
 Then you can post your own questions or snow reports...
Then you can post your own questions or snow reports...
davidof wrote:
... I commented that you cannot separate the model from its implementation - the thing that gives you some data - in the way it is used. Academically the implementation hasn't been peer reviewed and what little we know (ICL are keeping the March code secret) it is of unacceptable quality to be trusted..... In industry this would be a huge red flag for me.


Full disclosure: a fair bit of my own recent business has been the implementation, testing and calibration of statistical models for health care. You're on my lawn here.

Could you explain what you mean in a bit more detail, perhaps with some specific examples?

I'm happy to explain how "threads" work if you specify what you mean by that, although it's not specifically a modelling concept.
snow conditions
 After all it is free Go on u know u want to!
After all it is free Go on u know u want to!
davidof wrote:
rob@rar wrote:
@davidof, thanks. I'm not anywhere near qualified to comment on the technicalities you raise, but certainly agree with your point that these things should be open to scrutiny.


but the problem is political, Boris chose to go with Fergusson's predictions, that was his choice.

The code (at least in single thread mode) may produce the results that the model says it should. The predictions sound plausible.
Is the Imperial College modelling the only modelling that is being considered by SAGE?
ski holidays
 You'll get to see more forums and be part of the best ski club on the net.
You'll get to see more forums and be part of the best ski club on the net.
@davidof, some links to specific examples supporting your claims would be useful here.

I'm particularly interested in the following:
- Who is critiquing the ICL/ Ferguson team, and what are their (scientific) credentials?
- What is lacking in the open-source scientific resources the ICL team have made available on their website? (https://www.imperial.ac.uk/mrc-global-infectious-disease-analysis/covid-19/covid-19-scientific-resources/)...which doesn't look to me like 'ICL keeping the March code secret'

Also worth noting that the model they are using emerges from earlier work which was peer reviewed and published in Nature and PNAS.
snow conditions
 Ski the Net with snowHeads
Ski the Net with snowHeads
> - What is lacking in the open-source scientific resources the ICL team have made available on their website? (https://www.imperial.ac.uk/mrc-global-infectious-disease-analysis/covid-19/covid-19-scientific-resources/)...which doesn't look to me like 'ICL keeping the March code secret'

Show me the March code that Fergusson's claims were based on.

> Also worth noting that the model they are using emerges from earlier work which was peer reviewed and published in Nature and PNAS.

I'm not criticizing the model. You (and Rob) will have to look elsewhere for that.
ski holidays



Terms and conditions  Privacy Policy