cobafdek
-
Posts
525 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Posts posted by cobafdek
-
-
Another IYI not keeping up!
No, Taleb is being cheeky in an attempt to appear smart on twitter. If you don't assume what is called a Markov process, then you are left in a much worse (and less sensible) place.
Written like a true bot! Only bots live in Markov processes. The rest of us are in the real world.
The real world is non-ergodic. In a non-ergodic world, the so-called "fallacies" and "biases" of behavioral economics are not necessarily irrational.
-
Edit: Darn, broke my rule about not responding to nonsense dwellers.
Finally, we agree. You should stop posting here. Or at least you need to be more careful what you post.
Orthopa earlier outed you as a fake MD.
Now you're outed as a fake Taleb reader.
Nah, he's talking about people who underestimated him, but I'm not going to pretend to speak for him.
Taleb: "Everything after the 'but' contradicts what came before."
Sunk cost fallacy! Trump bag holders...
Taleb says the sunk cost fallacy isn't really a fallacy:
-
I'm not delusional on why Trump won and why he has a fair shot of winning again. I fully understand why his followers will follow him off a cliff. He is quite talented at what he does.
Funny that you left out the first part of Taleb's tweet:
Resist The Wuhan Lab & other theoriesSo basically, the conspiracy theories being peddled by Trump and eaten up by his followers. LOL. ;D
How convenient, but not surprised that a Trump supporter would try to mislead by distorting the truth.
You never asked me about the Wuhan lab theory. For the record, I don't believe it.
The plain meaning of Taleb's tweet is clear: when he says "IYIs on Trump," he's talking about you.
-
Bots, as well as anti-Trumpers deep in TDS, can't recognize negotiation tactics and diplomacy.
Chinese propaganda bots have infected CoBF. Examples:
If you have any facts to share on COVID19 or the response to the crisis, go for it. Otherwise, you're the Trump propaganda bot.
What is a China propaganda bot? Is this one?
“China has been working very hard to contain the Coronavirus,” Trump wrote in a post on Twitter. “The United States greatly appreciates their efforts and transparency. It will all work out well. In particular, on behalf of the American People, I want to thank President Xi!”
If you have to ask, you've failed the Turing test.
What ever happened to 'infidels'? What's up with the newfangled name calling?
Such clever propaganda -- all criticisms are just lies from the infidels. You are falling for it head over heels.
Yes, anyone who disagrees with them is a 'bot', 'traitor', 'not a patriot', etc.
Yet they're the ones who want to deny states suffering from crisis aid and they're the ones who put a sociopath in office that will "destroy the system".
They came over from the politics section section because they can't stand by while their dear leader is criticized for his terrible management of the pandemic!
Sunk cost fallacy! Trump bag holders...
Taleb on the above victims of TDS:
". . . victims of paranoia can find a narrative that impeccably matches all facts, people collectively fall prey to these shared delusions. 'Intelligent' people can be more vulnerable
Remember the Iraq WMDs or IYIs on Trump"
-
Bots, as well as anti-Trumpers deep in TDS, can't recognize negotiation tactics and diplomacy.
Chinese propaganda bots have infected CoBF. Examples:
If you have any facts to share on COVID19 or the response to the crisis, go for it. Otherwise, you're the Trump propaganda bot.
What is a China propaganda bot? Is this one?
“China has been working very hard to contain the Coronavirus,” Trump wrote in a post on Twitter. “The United States greatly appreciates their efforts and transparency. It will all work out well. In particular, on behalf of the American People, I want to thank President Xi!”
If you have to ask, you've failed the Turing test.
-
Chinese propaganda bots have infected CoBF. Examples:
https://www.snopes.com/fact-check/trump-admin-seizing-ppe/
"Under U.S. President Donald Trump, federal authorities are confiscating orders of personal protective equipment (PPE) from local governments amid the COVID-19 pandemic."
https://twitter.com/realDonaldTrump/status/1257741326650470400?s=20
Well run States should not be bailing out poorly run States, using CoronaVirus as the excuse! The elimination of Sanctuary Cities, Payroll Taxes, and perhaps Capital Gains Taxes, must be put on the table. Also lawsuit indemnification & business deductions for restaurants & ent.How to politicize a pandemic and basically blackmail people in doing your pet wishes while witholding the help that they need and deserve by being, y'know, part of the country.
What he calls bailouts are actually just being a single country.
When New Orleans or Mississippi or Florida gets hit by a huge hurricane and they get help, are they getting "bailed out" by the other states or is helping the worst hit places just the function of a government?
The most urban states are worse hit first by pandemics, that's biology, but he's claiming this is some kind of political thing to punish enemies. How sociopathic do you have to be as people are suffering and dying? Don't the urban coastal states pay a lot more to the federal government than many of the poorer states that aren't too badly hit yet? So they are supposed to pay more and get less in return? When he bailed out farmers when his tarrifs hurt them, that was fine because it was political allies, right?
Is he trying to break up the United States?
He must have read the weekend's headlines about his party deserting him, so he's showing support for eliminating cap gains, expanding business deduction limits, and cutting payroll taxes. And he's using coronavirus as the excuse to cover his tracks of obviously buying votes here, and covering his tracks of using that excuse by first accusing Democrats of using Coronavirus as an excuse.
I guess that's partly why Trump makes up reasons for not helping states when they ask for help. He's just too incompetent to actually do it even if he wanted to.
-
^Assessing the size of the invisible part of the iceberg is important. The California serology studies are interesting and need to be taken into account. However, if the validity of results is reduced by the same methodological flaws (selection bias, specificity bias underestimation {a specificity of 98% vs 99% can have a major effect on results} and extrapolation risk), results may be precise (values similar and pointing in the same direction) but also precisely wrong (not accurate). I think Mr. Maboussin discusses this noise-versus-signal issue for investments.
See below for the tests used (if i understood correctly) and look at specificity. There have been issues with tests rapidly going to the markets in this environment.
https://imgcdn.mckesson.com/CumulusWeb/Click_and_learn/Premier_Biotech_COVID19_Package_Insert.pdf
The manufacturers of that test report specificities of 99.2% for the IgM and 99.5% for the IgG, which sound excellent. But these are based on validating their tests by "Clinical Diagnosis/Confirmed," which presumably means symptomatic cases confirmed with a +RT-PCR. So that's the selection bias. It might be useful in a hospital setting. But when survey testing a larger and different population (asymptomatic or mildly symptomatic), they will get more false positives, which means lower specificity. It's interesting their package insert admits this much at the very end:
"Some cross reactivity was observed with samples positive for SARS-CoV antibody and Rheumatoid Factor. It is possible to cross-react with samples positive for MERS-CoV antibody. Positive results may be due to past or present infection with non-SARS-CoV-2 coronavirus strains, such as coronavirus HKU1, NL63, OC43, or 229E."
But they present no data on how large this cross-reactivity is. If large, specificity might be much lower. So it's probably not useful for community-wide testing for contact tracing, where a positive test will need to be confirmed by the RT-PCR.*
I fear a testing/tracking program would fail miserably. Not because its not the right thing to do or best thing to do but I feel it would border on downright impossible in a country like the US.
*One way it may be possible: Abbott's ID NOW COVID-19 test (a PCR test), according to my hospital which is about to deploy it (and according to Abbott - if they can be believed) has a very low false positive rate. It could therefore be used to confirm a positive serological test that has an unacceptably large false positivity. Abbott's rapid test, however, would have to be more widely available outside the hospital setting.
The McKesson test could still be useful for epidemiological purposes. Epidemiological models could use a range of false positive numbers to adjust their past models in order get better estimates of CFR and IFR.
I'm still waiting for Mt. Sinai's serological (ELISA) test (brought up in this thread last month https://www.medrxiv.org/content/10.1101/2020.03.17.20037713v2). I think they test for antibodies to different Covid-19 proteins. They claim no cross-reactivity to other non-Covid19 coronaviruses.
I give the S. Koreans a lot credit if this was the true reason it worked. God damn.
I bet the South Korean tests will be revealed also to have poor performance characteristics. So that a large factor for the relative success in that population was psychological. When they see a lot a weird "scientific" testing being done, combined with all that highly publicized spraying of the sidewalks/streets/buildings with whatever homeopathic fluid by men in hazmat suits, it's great propaganda to enforce quarantining.
-
Bill Maher, and the whole Trump-is-a-War-Criminal-He-did-Nothing-in-February crowd, get taken to the cleaners by Congressman Dan Crenshaw:
-
For the "nothing could be done anyway" and "please don't blame our leader" crowd:
Cafes bustled with customers, parks teemed with sunbathers, and the first Apple store to reopen outside China had lines snaking out the door as many South Koreans -- almost all wearing masks -- emerged from months of self-isolation.
Initially one of the hardest-hit with the second-highest number of cases globally, South Korea has managed to curb the spread without taking measures that were too severe. It didn’t require businesses to close or restrict travel.
While both the U.S. and South Korea confirmed their first virus cases around the same time in late January, the number of infections in the U.S. has swelled to more than 700,000 while Korea “flattened the curve” last month and cases have slowed to just over 10,000.
It's ok, it only cost our economy trillions of dollars. No big deal.
Let's not hold our federal gov't (CDC/FDA which botched testing and did not prepare back in January/Feb) accountable...
Actually, let's put all the blame on the lesser guys in federal government/state/local leaders (don't they run the FDA/CDC?), but leave nothing for the guy at the top. We need to provide him the asymmetry he always gets from his followers: no downside, only upside. Zero skin in the game "I take no responsibility" POTUS.
Should the President fire Fauci/Birx?
-
vacuum of leadership at the federal level
The Federal government is failing on the health response to the virus; no leadership or coordination etc. The result of this incompetence
Should the President replace Fauci/Birx?
At the end of the day the president owns the national response. If Fauci/Birx are providing bad advice then they should be replaced. If you watch 5 or more of Trump’s daily updates you will quickly realize who the problem is. Is that what a world class national response to a global pandemic looks like?
ARE they providing bad advice?
-
vacuum of leadership at the federal level
The Federal government is failing on the health response to the virus; no leadership or coordination etc. The result of this incompetence
Should the President replace Fauci/Birx?
-
I'm trying to ask this uncontroversially mostly medical professionals here: assuming a gradual relaxation of the lockdown with no vaccine, no treatment, and no (or minimal) testing/tracking. Would you say going out is acceptable or too risky? Let's assume a medium-high risk state like MA. Let's assume not super vulnerable person.
If we want to talk concrete "going out" categories, let's say going to parks, general shopping, meeting friends/family, going to office/work, going to a restaurant (I tried to order this from least risk to most risk).
Outcome of "severe" infection sounds very scary. That has to be balanced with infection risks though.
I know this is a bit theoretical and uncertain, but since there's a talk of "relaxation" even in NY state, maybe this could be useful.
I could open a new topic... but probably not worth it.
Thanks
Today, yes, I think the risk is acceptable, based on general reports of curve flattening and general lack of overwhelmed local ERs and ICUs.
The "re-opening" of local economies should be conditional on:
1. universal mask usage
2. maintaining 6-ft social distancing
3. no large crowds
This can allow many businesses, restaurants, maybe some schools to re-open.
The decision regarding timing, pace, and extent of re-opening will be left to state governors and local authorities, who can fine-tune the above conditions, based on local factors. If a surge of serious cases start to show up in ERs and ICUs, they'll have to clamp down again.
The federal government level can mostly give permission to states and localities to open up around April 30 (maybe even before), when they feel ready. They can veto crazy decisions that might happen in some regional southern areas, such as large church services, movie theaters, sports arenas.
As you say, there is no good medical treatment/vaccine, and testing has been disappointing. (Whenever you rush out with new tests for a new disease, being unable to evaluate accuracy and reliability systematically, we really can't trust the results.) But you don't need testing or treatment in new pandemics, since the only effective measure is various levels of quarantine and travel restrictions. (If South Korea had no testing but had only lockdowns, they would have still been fine.)
Today, I think the risk of economic recession/depression (deaths, suicides, depressions, lack of confidence in authorities) is greater than the risk of swamping the medical system, especially if the heavy lockdown extends beyond April 30. I think this feeling is widespread, and is percolating from the bottom-up. Any heavy handed top-down governmental restrictions will be answered by spontaneous bottom-up rebellion - so in a way, I think your question is moot.
Improvement in testing should go on, but their utility will primarily be in retrospective analysis for future outbreaks. We don't need precise knowledge from test results to know what to do now (again, the only thing to do is isolation with masks and some form of quarantine). And even if a vaccine is developed, I don't have confidence that it will be effective or safe enough, especially if it comes earlier rather than later.
-
Anthony Fauci MD. And war criminal, behind the War Criminal. Fauci has effectively joined the campaign to re-elect the President.
"Fauci said that the 'first and only time' that he and coronavirus response coordinator Dr. Deborah Birx talked to Trump about 'shutdown'-like mitigation policies, 'the president listened to the recommendation and went to the mitigation.'”
"When Fauci and Birx went to Trump a second time advising him to extend the White House’s social-distancing guidelines through the end of April, Trump 'went with the health recommendations,' Fauci said."
Anyone surprised that Fauci's statements (above) were left out of the Washington Post's account of today's briefing?:
-
So we are now at the stage of "nobody else could have done anything anyway" and "look at all the things he is doing now". Expect this to persist till November.
Some of us work hard to develop foresight (which you can verify by looking back on this thread: worrying about virus/testing in late Feb when your boy was calling it "a hoax" and "like the Flu" or your congressmen who had foresight to dump stocks, and when we don't achieve foresight, we ought to then use hindsight (acknowledging & learning from our mistakes), but some people instead choose no sight (willful blindness).
We know this guy and his supporters' MO. Nothing is his fault (it was Sessions, Flynn, Bolton, etc etc a.k.a. the buck never stops at me) and "I can shoot a man on 5th ave and I wouldn't lose voters". I much prefer the "you can fool some of the people all of the time" quip which is credited to a much greater President (though doubtful he said it).
For the record, I fault the administration for:
1. Listening to Fauci which likely contributed to delaying the China travel ban back in January.
2. Over-reliance on models.
3. Not mandating, or at least strongly recommending, masks or similar facial covering.
Netanyahu is a friend of the President. You can bet the administration is considering (or at least should be) adopting his idea for cellphone tracking and stiff penalties. I'll go further and propose similar penalties for not wearing a mask.
Let's say you're part of the task force discussion. Where do you come down on this issue?
-
Looks like thousands of American citizens are going to be crucified at the alter of one persons narcissism. A "war time" president who should be facing war criminal charges of epic incompetence.
https://www.washingtonpost.com/world/2020/03/31/coronavirus-latest-news/
The White House coronavirus task force on Tuesday presented a grim picture of where the U.S. could be heading over the next couple of months, even with interventions like physical distancing. The task force projects 100,000 to 240,000 deaths from the virus, with mitigation.
Deborah Birx and Anthony S. Fauci, the leaders of the task force, emphasized that although the projections were likely based on the data that they have seen from the hardest hit locations so far, they were hopeful that they could prevent such a high number of deaths.
“Whenever you’re having an effect, it’s not time to take your foot off the accelerator, and on the brake, but to just press it down on the accelerator,” Fauci said of the mitigation efforts. “And that’s what I hope. And I know that we can that do over the next 30 days.”
Spot on. The lack of basic understanding shown in early stages of this is beyond belief. I still believe they are downplaying the true possible numbers here. Two more weeks and we will know, by tomorrow US will likely have 200K+ infections.
Question for the Hindsight Geniuses on this thread:
Which one(s) of the anti-pandemic measures in the following article should the President have done in January? Which ones should he do now?
https://www.nytimes.com/2020/03/30/world/europe/coronavirus-governments-power.html
I'm not interested in argument. I asked a serious question to get people on record. So far responses from those I'm addressing have been non-existent and have shown the expected cognitive dissonance.
I'll go first. I'm in favor of this:
"Israel’s prime minister, Benjamin Netanyahu, has authorized his country’s internal security agency to track citizens using a secret trove of cellphone data developed for counterterrorism. By tracing people’s movements, the government can punish those who defy isolation orders with up to six months in prison."
-
Looks like thousands of American citizens are going to be crucified at the alter of one persons narcissism. A "war time" president who should be facing war criminal charges of epic incompetence.
https://www.washingtonpost.com/world/2020/03/31/coronavirus-latest-news/
The White House coronavirus task force on Tuesday presented a grim picture of where the U.S. could be heading over the next couple of months, even with interventions like physical distancing. The task force projects 100,000 to 240,000 deaths from the virus, with mitigation.
Deborah Birx and Anthony S. Fauci, the leaders of the task force, emphasized that although the projections were likely based on the data that they have seen from the hardest hit locations so far, they were hopeful that they could prevent such a high number of deaths.
“Whenever you’re having an effect, it’s not time to take your foot off the accelerator, and on the brake, but to just press it down on the accelerator,” Fauci said of the mitigation efforts. “And that’s what I hope. And I know that we can that do over the next 30 days.”
Spot on. The lack of basic understanding shown in early stages of this is beyond belief. I still believe they are downplaying the true possible numbers here. Two more weeks and we will know, by tomorrow US will likely have 200K+ infections.
Question for the Hindsight Geniuses on this thread:
Which one(s) of the anti-pandemic measures in the following article should the President have done in January? Which ones should he do now?
https://www.nytimes.com/2020/03/30/world/europe/coronavirus-governments-power.html
-
Same with Obama's 800B infrastructure plan of which like, 60B? was actually spent on projects.
Same story, new decade.
Pessimistic, naive empiricism.
-
Anybody opposed to this idea?
Nobody has ever been opposed to infrastructure spending. It just never happens.
https://twitter.com/realDonaldTrump/status/1245000074167541761
-
Anybody opposed to this idea?
-
I have nothing but respect and admiration for this gentleman. Given his work conditions and boss, he still tries his best to have utmost integrity.
Consider this from January 26, 5 days before the President ordered the ban on travel from China:
Feel any different?
-
We're going to get serological results quicker than I thought. Recent past CFR numbers will be coming down.
A tweet from yesterday:
was updated today:
Definitely going to be watching this. Do you know by chance is this an independent company? What is driving this goverment? His profile is lackluster.
I don't know who he is or what company is testing. I gather that Friedberg is an angel investor in venture capital.
There must be dozens of start-ups, not just in the USA, that are racing with the serological testing. The Mount Sinai lab that published their results last week (https://www.medrxiv.org/content/10.1101/2020.03.17.20037713v1) has a website where competing labs can order the ingredients and do the testing themselves https://labs.icahn.mssm.edu/krammerlab/covid-19/
Things are moving very fast. Good news.
-
We're going to get serological results quicker than I thought. Recent past CFR numbers will be coming down.
A tweet from yesterday:
was updated today:
-
Definition of COVID-19–Related Deaths
A second possible explanation for the high Italian case-fatality rate may be how COVID-19–related deaths are identified in Italy.
Case-fatality statistics in Italy are based on defining COVID-19–related deaths as those occurring in patients who test positive for SARS-CoV-2 via RT-PCR, independently from preexisting diseases that may have caused death. This method was selected because clear criteria for the definition of COVID-19–related deaths is not available.
.......
Electing to define death from COVID-19 in this way may have resulted in an overestimation of the case-fatality rate.
.........
The presence of these comorbidities might have increased the risk of mortality independent of COVID-19 infection.
This is what I should have highlighted as being misleading.
Couldnt they have used XRay or atleast breathing problems as a criteria?
A 80 year old heart attack patient with clear lungs needs to be counted as Covid 19 death because of a positive PCR test?
It's a great question. The spectrum of Covid-19 illnesses has yet to be defined. For all we know now, Covid-19 could have caused that heart attack. Until the time that standardized disease definitions are decided upon, and at this early stage of our knowledge, a positive test is all we can use to report CFR of Covid-19.
A positive influenza test in an 80 year old heart attack patient with clear lungs who dies is an influenza death, with the other comorbidies listed as contributory.
-
My point:
Covid-19 MUST be included in the cause of death.
It's my response to
Nobody was disagreeing with that.
Let me clarify:
Definition of COVID-19–Related Deaths
A second possible explanation for the high Italian case-fatality rate may be how COVID-19–related deaths are identified in Italy.
Case-fatality statistics in Italy are based on defining COVID-19–related deaths as those occurring in patients who test positive for SARS-CoV-2 via RT-PCR, independently from preexisting diseases that may have caused death. This method was selected because clear criteria for the definition of COVID-19–related deaths is not available.
.......
Electing to define death from COVID-19 in this way may have resulted in an overestimation of the case-fatality rate.
.........
The presence of these comorbidities might have increased the risk of mortality independent of COVID-19 infection.
This is what I should have highlighted as being misleading.
Coronavirus
in General Discussion
Posted
Uh oh, anti-Trumpers:
https://uk.reuters.com/article/uk-health-coronavirus-who-idUKKBN26U1ZW
Trump wins.