SLU & NCAA Corona Virus Discussion

[Old guy]

4 hours ago, billiken_roy said:

deal with it?  we have over 35,000 deaths in america every year rom the flu.  that is dealing with it?

again, i am not denying there is a chinese flu virus that is causing problems.   i just want to know why the hysteria and panic for this virus when we have had and do have very serious problems with other viruses every year and nothing like this current panic has ever occurred before.   

Roy, my friend, no one lives forever. There is a cause for every  single death that happens, whether it is a car accident, or a virus, or cancer, or whatever. Even when all that is possible to do is done people still die. We are all aware of this and accept it. The thing that makes this Covid different is its newness. And yes, absolutely true in this regard, there may be all kinds of discussions, recriminations and accusations as to "who did what and with which and to whom" afterwards. These issues are unlikely to be resolved satisfactorily.

[Old guy]

Just now, courtside said:

 You referenced the Imperial College Model and made conclusions from which the author of it disagrees.

Did it make you feel good saying that? You are merely quoting a "higher authority." Again, believe what you want.

[BilliesBy40]

56 minutes ago, billiken_roy said:

You are nuts.   we have shut down america and are knocking on the door of a major economic depression.  that is panic.   the media is now being destroyed every night by president trump.   i love that the tables are turned on the fake news msm.  finally america that only watches fake news is seeing for themselves how dishonest and one sided the media has become.  

What exactly is fake in the media’s reporting of this pandemic?

[brianstl]

If you really want to understand the possible risk with dealing with what we are currently dealing follow Nassim Nicholas Taleb. Should we do everything he says? I don’t think so, but he understands the fat tail risk of these types of situations better than most of the experts you hear talking about this and was far ahead of any of them in understanding the problems we faced with this virus.  
 

Here is his take on the British models.

https://amp.theguardian.com/commentisfree/2020/mar/25/uk-coronavirus-policy-scientific-dominic-cummings?CMP=share_btn_tw&__twitter_impression=true

 

[courtside]

25 minutes ago, Old guy said:

Did it make you feel good saying that? You are merely quoting a "higher authority." Again, believe what you want.

It isn't about what I believe or don't believe. It was about you posting misinformation that was corrected by the person of whom whose information you used. Did I feel good posting it? I felt indifferent because it's about you and him.

[billiken_roy]

1 hour ago, Old guy said:

Roy, my friend, no one lives forever. There is a cause for every  single death that happens, whether it is a car accident, or a virus, or cancer, or whatever. Even when all that is possible to do is done people still die. We are all aware of this and accept it. The thing that makes this Covid different is its newness. And yes, absolutely true in this regard, there may be all kinds of discussions, recriminations and accusations as to "who did what and with which and to whom" afterwards. These issues are unlikely to be resolved satisfactorily.

All the flus and cancers and ailments that came before this were new at one time as well.  Why did we not have a world shut down panic then as well?

[GBL_Bills]

On 3/27/2020 at 8:57 AM, billiken_roy said:

in my view, it started in china, it is another strain of the flu and it is a virus.   

Didn’t want to read through 6 pages of politicking, so don’t know if this was corrected or not. 

Regardless of what you call it, COVID-19 is not just a flu strain. Coronaviruses are an entirely different family of viruses than influenza. Calling it “just another strain of flu” could lead people to think that there is potential for crossover therapy/vaccines. 

[rgbilliken]

1 hour ago, brianstl said:

If you really want to understand the possible risk with dealing with what we are currently dealing follow Nassim Nicholas Taleb. Should we do everything he says? I don’t think so, but he understands the fat tail risk of these types of situations better than most of the experts you hear talking about this and was far ahead of any of them in understanding the problems we faced with this virus.  
 

Here is his take on the British models.

https://amp.theguardian.com/commentisfree/2020/mar/25/uk-coronavirus-policy-scientific-dominic-cummings?CMP=share_btn_tw&__twitter_impression=true

 

Taleb’s book The Black Swan is great. 

[Old guy]

1 hour ago, courtside said:

It isn't about what I believe or don't believe. It was about you posting misinformation that was corrected by the person of whom whose information you used. Did I feel good posting it? I felt indifferent because it's about you and him.

The fact remains that you cannot look a the model's assumptions as I did and declare them invalid. Therefore you have to substitute  the knowledge you do not have for belief . Believe what you wish.

 

 

[Old guy]

33 minutes ago, billiken_roy said:

All the flus and cancers and ailments that came before this were new at one time as well.  Why did we not have a world shut down panic then as well?

Roy, you are talking about the  history  of medicine here. For example, are you aware that the first description of a heart attack (myocardial infarction) was not developed until the 1920s. Before that heart attacks surely existed, they have existed forever since very few things are new, but they just could not be either detected as something "new" or defined until the EKG was developed after WWI. Infectious diseases have been here forever, but they were thought to  be due to humans displeasing God and being punished for this. It was not until Pasteur and Koch that the veil started to  be lifted. Understanding and comprehension of something does not happen automatically once that something if found to exist. Our knowledge of Alzheimers as ample as it has become does not allow for control or prevention of the disease and so on and so forth. For all we know, there have been epidemics of Covid before, but they were never detected as something else that was different.

Panic is not something that happens regularly when a new disease is discovered, it happens when it becomes a way to achieve a goal, whether the goal is to sell papers, or whatever it may be. We do have a panic now, but that is an entirely different animal from the disease Covid-19

[Sheltiedave]

“As we said, when one deals with deep uncertainty, both governance and precaution require us to hedge for the worst. While risk-taking is a business that is left to individuals, collective safety and systemic risk are the business of the state. Failing that mandate of prudence by gambling with the lives of citizens is a professional wrongdoing that extends beyond academic mistake; it is a violation of the ethics of governing.”
 

Brian, Taleb and his partner do a good job of framing the ins and outs of risk modeling, but they fail in part as they enter into medical modeling. It is radically different than engineering modeling, where there only three to six parameters beyond material composition that will encompass a robust modeling.

Courtside, tossing out your animosity for Old Guy, I am absolutely certain he has serious, valid, and professionally informed opinions about how the British projections are skewed, and from the initial huge numbers that were advanced, I would wager he did not bound his model in the proper conventional fashion.

If he led the published paper with an unbounded model outside of accepted practice, he could easily go off the reservation. Let’s do a quick model for England.

England has 66.4 million people. Let’s set Rought at 2.3, 1000 people infected, each generation is 3 days, how many people would be infected after 45 days, how many people would die based on a 3% mortality rate, no mitigation?

y = m Ro ^T/t

y = 1000(2.3)^45/3 = 1000(2.3)^15

y =1000(266,635) = 266,635,000

total infected defaults to 66.4 million

66.4 million x 0.03 = 1,992,000 deaths

[Old guy]

That is good Sheltie Dave,  you have created a brief model (to illustrate your point) that means absolutely nothing. This can and does happen with models. I once read a paper from a colleague who did extensive modelling of cholesterol, triglycerides, and mortality. I think his paper described over a dozen different models he had developed for this purpose. Not one of them produced anywhere close to the actual mortality experience results available at the time. Therefore he chose one of them, the one that fitted his way of thinking better (this is exactly what he said in his paper) and proclaimed it to be able to predict future mortality that would be expected to occur in our changed modern world. Key words in his rationale being changed and modern.

BS is BS.

[Sheltiedave]

1 minute ago, Old guy said:

That is good Sheltie Dave,  you have created a brief model (to illustrate your point) that means absolutely nothing. This can and does happen with models. I once read a paper from a colleague who did extensive modelling of cholesterol, triglycerides, and mortality. I think his paper described over a dozen different models he had developed for this purpose. Not one of them produced anywhere close to the actual mortality experience results available at the time. Therefore he chose one of them, the one that fitted his way of thinking better (this is exactly what he said in his paper) and proclaimed it to be able to predict future mortality that would be expected to occur in our changed modern world. Key words in his rationale being changed and modern.

BS is BS.

Old Guy, I know it means nothing, because it doesn’t capture anything beyond a purely exponential rate, which all the graphs show only track for a few generations. It doesn’t track at all for reduced transmission, nor for isolation, nor for many other factors that rapidly enter, like reduction of population pool available for infection.

it does show we can slow it by social distancing, as that greatly expands the generation time factor

That is why we need to listen to the Infectious Disease, Pandemic, Virologists, and Epidemiologists like you that are THE experts here. We understand a little, you spent your life working this field.

[Old guy]

Sheltiedave, I am not putting you down. I intended to put down the guy with cholesterol, triglycerides, and mortality, but I guess it did not come out correctly. Sorry about this.

[Basketbill]

Old Guy,  can you be specific about the errors in the first Paper from the Imperial College.   I have seen both and some have said that the first was filled with errors and it had to be revised.   I would not describe that as such,  this paper release on the 26th has just a grim an estimate, " if nothin is done or unmitigated".   This second paper is more of a revision due to additional data, new considerations and an expansion of predications to cover not just the UK but also many other countries with a focus on the higher GDP countries.  

to say that the estimate is reduced is wrong,  I see no reduction at all.  Perhaps I am missing something in the data. 

                                                                                                                                                       Total Pop.       cases.            deaths

[moytoy12]

Meanwhile, in Oklahoma, my dad tells me rural restaurants are still serving dine in meals as of last night. The stubborn among us might be impacted the most. Oklahoma is going to be decimated by this virus.

Also, I know it’s hard to understand for some, but a pandemic can be worse than a seasonal disorder. If you don’t understand how or why, you should probably just shut up.

[Old guy]

1 hour ago, Basketbill said:

Old Guy,  can you be specific about the errors in the first Paper from the Imperial College.   I have seen both and some have said that the first was filled with errors and it had to be revised.   I would not describe that as such,  this paper release on the 26th has just a grim an estimate, " if nothin is done or unmitigated".   This second paper is more of a revision due to additional data, new considerations and an expansion of predications to cover not just the UK but also many other countries with a focus on the higher GDP countries.  

to say that the estimate is reduced is wrong,  I see no reduction at all.  Perhaps I am missing something in the data. 

                                                                                                                                                       Total Pop.       cases.            deaths

Yes, I can. The original Imperial College article was published here:

https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College-COVID19-NPI-modelling-16-03-2020.pdf

My analysis of this article was published in this thread, page 25, March 18th, this is  what I posted at that time (you may check the original post, it is still there):

"This is a well done article. It is basically a study on the effect of mitigation techniques applied to a nationwide situation. They use A LOT of assumptions in their work, which is entirely OK because the direction of the article is to  show variation in the total mortality when different mitigation methods are used and the timing of their adoption. So, it really does not matter how many deaths actually happen as long as the results of the mitigation techniques can be applied uniformly and the differences resulting from their application calculated. They assume that 85% of the total populations of the US and UK are expected to contract the disease. Assume the incubation period will be 5.1 days, that the infected people will be capable of transmission from 122 hours prior to developing symptoms to 4.6 days after becoming symptomatic. They assume that individuals that are asymptomatic are 50% more infectious than they are after the onset of symptoms. They assume the R0 is 2.4. They assume the infection was seeded in each country early January 2020 and has grown exponentially ever since. They assume 50% of the cases will remain undetected and develop either no symptoms or very mild symptoms. They assume (from Chinese data) that only 4.4%  of the confirmed infections will require hospitalization, and that 1/3 of those hospitalized will  require intensive care. They assume (from Chinese data) that 1/2 of those requiring intensive care will die. They also calculate an IFR (infection fatality ratio) of 0.9. And here comes the cherry on top of the whipped cream. Since the number of deaths after hospitalization and failed ventilatory efforts are not enough to reach the calculated IFR, they add a round up number of additional undocumented deaths, primarily among the very old, to a level sufficient to fit the assumed IFR of 0.9%.

I hope you read all of this because it is all a model based upon assumptions that are required to bring us to the  grand total of over 2 M deaths expected in the US. ALL THE ASSUMPTIONS IN THIS MODEL MUST BE MET TO REACH THE GRAND TOTAL OF 2 MILLION DEATHS IN THE US. This model is OK to study the results of mitigation interventions taken at different times, IT IS NOT OK AS A PREDICTOR OF TOTAL US DEATHS for the reasons described above.

There will be all kinds of articles done in this  manner which will be equally invalid as a predictor for US deaths this year. The assumptions made in this model are almost biblical, and this one begat, these many others, and each of the  others begat even more. The model ends with a fudge factor when the  number of deaths do not fit the assumed IFR.

Death is serious business, this model is not valid mortality prediction tool, it is just a way to compare and quantify the value of mitigation strategies. For this purpose this article has value.

I stick to my prediction of less than 26,000 deaths in the US when all is said and done. Are you aware that China which has 5X our  population and an inadequate health care system for a large portion of their population (traditional Chinese medicine) had reported only 3000 deaths by March 15th? Can the US with a better and more uniform health care system coverage and 20% of the Chinese population be seriously expected to reach 2,000,000 plus deaths? NO WAY.

Happy St. Patrick's, one day late, to everyone."

The assumptions taken in this article may have been chosen to magnify the effects that certain forms of mitigation and the timing of the  application of such mitigation measures would have upon the model portrayed. This is valid. The model used by itself and the assumptions used are not valid, particularly not in regards of the IFR chosen and the addition of an undisclosed but presumably large number of documented deaths to fill the deficit produced by using their assumptions and the IFR chosen. The article's model predicted 500,000 deaths for the UK and 2.2 M deaths for the U>

The chief author of the  article, Dr. Ferguson, which was taken by the media and proclaimed as the "immense danger we face" in this epidemic, was called to testify in front of the British Parliament about the expected mortality faced by the UK. There should be no doubt in anyone's minds that testimony given in front of the legislature HAS TO BE CORRECT and be supported by the available data at the time. Dr. Ferguson could not confirm the model's prediction of 500,000 deaths in the UK due to this epidemic. He ratcheted down the estimated number of deaths in the UK to 20,000. This is a verification of the invalidity of this particular model to accurately predict mortality in the UK and just as likely everywhere else. This says it all. A particular problem with this model is, as I pointed out, the "undocumented deaths" used to validate the chosen IFR.

This is an epidemic that suffers from panic inducing statements by experts and by media pronouncements out of proportion to reality.

The virus is real, the epidemic is real, the fact that the virus will kill people is real, the panic is real, and the loss of jobs is also real. My own estimate which is not based upon a model, but rather is just based upon experience is for less than 26,000 deaths in the US. I stand by it. I have no idea whatsoever what may happen with the economy, or with the media coverage, or whether or not the panic will be effectively controlled.

I hope this provides adequate documentation, including the paper of the model used by Ferguson and his group.

I have not accessed or reviewed the paper released 3/26. If you have it please provide a link.

[courtside]

2 hours ago, Old guy said:

The fact remains that you cannot look a the model's assumptions as I did and declare them invalid. Therefore you have to substitute  the knowledge you do not have for belief . Believe what you wish.

 

It's his model. He clarified and explained his model because some people such as yourself were drawing incorrect conclusions from it.  Now you repeatedly told me to believe what I wish. Well, why not believe the gentleman who's model you posted? You did until the conclusion isn't what you wanted so now you don't. Got it.

[Old guy]

@Basketbill. Maybe I was not specific enough, but a series of assumptions as those made in the model for the paper referred to above are rarely if ever met in reality. Minor variations  between assumed parameters and what is found in reality can cascade and cause major cumulative errors overall. Their use of an IFR of 0.9%, particularly when found to result in a wide underestimation of the number of deaths calculated on the basis of unsuccessful therapy and requiring use of "undocumented deaths" is a major problem with this article.

[Old guy]

9 minutes ago, courtside said:

It's his model. He clarified and explained his model because some people such as yourself were drawing incorrect conclusions from it.  Now you repeatedly told me to believe what I wish. Well, why not believe the gentleman who's model you posted? You did until the conclusion isn't what you wanted so now you don't. Got it.

Of course courtside, you are free to believe what you wish. The proof will be in the total number of deaths on the other side of the epidemic.

[courtside]

4 minutes ago, Old guy said:

Of course courtside, you are free to believe what you wish. The proof will be in the total number of deaths on the other side of the epidemic.

I don't think you understand. I do not care about the Imperial model. Your changed conclusion is incorrect according the author. My source was the author in his own words. I am not involved in your dispute with Ferguson. Take it up with him. What I do care about is misrepresented information. Many other aspects of COVID-19 are of more interest to me.

[Sheltiedave]

Courtside, correct me if I am wrong, but didn’t Old Guy provide his estimate well before the Imperial College paper was released?

Didn't the professor testify he thought maybe 20,000 deaths would be a better target model than 500,000 deaths?

Reads to me like OG has a solid projection of US mortalities, and is attuned to the Brit’s errors.

There where a number of if this, then this columns in his paper. Old Guy disagrees with one of his pivot linchpins, and that critical linchpin throws off his estimate by a factor of 25.

Heck, I did nothing more than a basic exponential equation, calculation on the envelope, knowing nothing about the wrinkles, and only missed his bad upper bound by a factor of 4.

 

[Old guy]

26 minutes ago, courtside said:

I don't think you understand. I do not care about the Imperial model. Your changed conclusion is incorrect according the author. My source was the author in his own words. I am not involved in your dispute with Ferguson. Take it up with him. What I do care about is misrepresented information. Many other aspects of COVID-19 are of more interest to me.

Just tell me where the misrepresentation is at. Can you do this?

[courtside]

43 minutes ago, Old guy said:

Just tell me where the misrepresentation is at. Can you do this?

According to Neil Ferguson (not me) Ferguson did not change projection in number of deaths as you stated in your post. My source is him, as in the guy involved with the exact models you referenced. This will be my last post on the subject. Read his tweets which I kindly posted in my first response and start there and work backwards. 

Again of all the wide variety of things associated with COVID-19, this is one of which my interest level is small. Your post initial post was not accurate according to Neil Ferguson. I recommend you strike up a conversation with him movinh forward, as, I am not involved.

[brianstl]

54 minutes ago, courtside said:

According to Neil Ferguson (not me) Ferguson did not change projection in number of deaths as you stated in your post. My source is him, as in the guy involved with the exact models you referenced. This will be my last post on the subject. Read his tweets which I kindly posted in my first response and start there and work backwards. 

Again of all the wide variety of things associated with COVID-19, this is one of which my interest level is small. Your post initial post was not accurate according to Neil Ferguson. I recommend you strike up a conversation with him movinh forward, as, I am not involved.

In Old Guy’s defense, he isn’t some MBM mouthing off on here.  This is the stuff he did for a living.  Also, many in their field have took issue with Ferguson’s work.