Extreme events and how to live with them by Nassim Nicholas Taleb
Professor Nassim Nicholas Taleb, New York
Distributions that are dominated by extremes and tail events require a completely different way of thinking. We provide a classification and show where conventional statistical tools fail, such as the conventional law of large numbers. We show how robust statistics is not robust at all; how frequency-based forecasting fails and how past averages misrepresent future ones. We show implications for decision-making in the real world and what modifications are required. Ironically they are often easier to work with. See www.fooledbyrandomness.com/Fat... for more details and papers.
Biography
Nassim Nicholas Taleb spent 21 years as a risk taker before becoming a researcher in philosophical, mathematical and (mostly) practical problems with probability. Taleb is the author of a multivolume essay, the Incerto (The Black Swan, Fooled by Randomness, and Antifragile) covering broad facets of uncertainty. It has been translated into 36 languages. In addition to his trader life, Taleb has also published, as a backup of the Incerto, more than 45 scholarly papers in statistical physics, statistics, philosophy, ethics, economics, international affairs, and quantitative finance, all around the notion of risk and probability. He spent time as a professional researcher (Distinguished Professor of Risk Engineering at NYU ’s School of Engineering and Dean’s Professor at U. Mass Amherst). His current focus is on the properties of systems that can handle disorder (“antifragile”). Taleb refuses all honors and anything that “turns knowledge into a spectator sport”.
This talk is part of the Darwin College Lecture Series series.
Taleb is brilliant. I have studied and practiced his work for 7 years. He is a Genius, and more importantly, his work is Lindy.
Any key takeaways? I have also been through much of his work but I'm not half as smart as I should be haha.
Lindy? What does that mean?
Aaron McKeehan Google The Lindy Effect.
@@TheNeojade Thanks, thats what I needed.
@@IsaacWendt Normal distribution and central tendency are phnomena of mathematics, not of the real world. The real world is stochastic (hence the Lindy effect), random, chaotic, and importantly from the point of view of Taleb's teaching, fat-tailed. Also heteroscedastic. en.wikipedia.org/wiki/Heteroscedasticity Back when the totalizer took only 85% at the track, i.e. the government plus the Jockey Club got by on 15%, there were many tracks (i.e. betting crowds) where you could make money automatically, given deep pockets, by betting the longest and shortest odds. That's fat tailed events. Today most tracks are taking 17%, so you can only win through discipline, hard work and good judgment. So sad!
I love how Nassim laughs at all of his own jokes. They are brilliant, I don't understand why all the people in the audience aren't laughing?
They are brain dead from thier Neoliberial, Friedman, Marx university education.
@Michael Auliffe You really got everyone included there haha. Not saying: I agree with you... just that you got a dig in at everyone from the right to left and vice versa! 😄
@@michaelayliffe7238 It's best to switch off your autopilot before you try to return to earth.
Because, being mainly English, they're too polite - or thick.
They don't like him.
My mentor ! The best in the business ...keep teaching us Master!
This video is just another sign of our addiction to probabilities first, find the data later. Everyone is so motivated to think of what might be the next big thing just so they can have something with their name on it.
Thank you very much indeed for posting this.
So you couldn't work it out for yourself.
This is why I invest and don't speculate. I'd be competing against guys like him.
he does not speculate
@@danielracovitan9779 sure he does - thats what trading is - he buys things he thinks will go up or down in value, and makes a ton of money because he is way way better at it than almost everyone else
@@danielracovitan9779 At various times he's suggested to keep 80-93% of your assets rock solid safe and to speculate if you wish with the rest.
Nassim Rocks. Love the Matrix wardrobe.
excellent introduction - the best i've seen on YT. She totally steals his thunder with her closing remarks too :)
Guess you are knew here.
That's not hard.
This is a great talk, because Taleb keeps it simple. Even his simple ideas tend to fly right over people's heads.
Oh, ha ha.
He ain’t lying. Very Lindy talk
mycroftc 1 week ago NNT begins around 6:18 time mark.
Underrated comment
- Recognize the difference between Mediocristan and Extremistan (0:44) - Understand the limitations of Gaussian statistics (0:58) - Learn about the importance of fat-tailed distributions (9:41) - Avoid using standard deviation and variance as measures under fat tails (22:25) - Apply maximum likelihood methods for parameter estimation (22:59) - Be aware of the gap between confirmatory and disconfirmatory evidence in fat-tailed domains (23:07) - Understand the concept of ergodicity and time probability versus ensemble probability (49:18) - Use heuristics and stress testing to assess fragility to tail events (1:01:11)
NNT begins around 6:18 time mark.
Thanks.
Taleb is no doubt the most profound thinker of maybe the last 400 years
Listen to Osho
Ayn Rand is
58:00 time probability vs ensemble probability
"that Swedish medal" LOL
What seems to be the necessary add-on is the severity of the results of the lower-probability events. For example, when calculating armor penetration the value used in many documents is the "V50" ballistic limit velocity (BLV), which is the estimated striking velocity where half of the enemy projectiles will punch through the armor and do most if not all lethal effects on your ship/vehicle (AND YOU, of course). However, is that really the value that you want when one hit that penetrates can "ruin your whole day"? I would think not and that somewhere around the V10-V25 BLV would be what would interest most military personnel behind said armor plate if they were in a battle, since it gives them enough chance of surviving a hit and retaliating so that the enemy does not get another chance to shoot at them that is meaningful in a battle. This also includes the time factor of having additive multiple events (enemy hits on you), which obviously are best when the enemy gets the fewest shots.
Hi, I wanted to comment and would love to enter discussion about this idea with you. I think you need to broaden your thinking in terms of how people chose material for armor manufacturing. Undestanding this, I feel, is what will help you see that statistics simply aims to mitigate the problem coming from v50 that you pointed out. Statistics isn't the failure of v50 as a metric, it's about the conscientious use of v50 as a metric to chose armor material. V50 is a metric to describe stability of different armor materials under stress. But I'm willing to bet it's not the only one used in optimizing choice of material used for armor. Instead, the MBAs (incellectual yet idiot) making the decision of which material they want use linear combinations of different metrics (price, weight, manufacturability, etc. in addition to V50). EM- algorithm provides the scalar weights for those properties. And those linear combinations which surpass the estimated stress of battle prove to be good materials. It is in the decision of the weight (think weighted average) that the tail risks taleb discusses come into the picture (or, rather, don't show up obviously). V50 is not the problem- to fix V50 is wayyyy harder than simply fixing how you consider different properties (price, weight, manufacturability, v50, etc.). In other words, picking what's important to you, the soldier, (perhaps a SEAL team will be willing to sacrifice penetrability for weight, i'm not sure i don't have military background), is where statistics can help. Just guessing is stupid (not efficient), but using statistics and calculus to optimize the choice of linear combinations resulting in the best material for battle is where one needs to also consider tail events. e.g. weight shouldn't be too much of a concern because if you get shot, you're fucked, but if you sweat more or are more tired after battle, you can still recover- most people would be tired over dead). It makes sense in something like fighting. In banking, it's all numbers on a screen. People lose sight of the assumptions they make (it's much easier to rationalize the prospect of dying by bullet wound v.s. weight of armor over the choice to lower interest rates to "nudge" the economy. Thus, we see entire policies (coming out of both parties) which make grave assumptions about what people can fix. It's not always easy to optimize something like battle. People will get shot.
@@HarryPotter-kd3bh My comments concerning the various "Vnn" values ("nn" being the average statistical probability of the armor working -- barring hits on cracks or other such "low-ball" things) were not for manufacturing tests, where V50 is probably the best measure of the protection from a given piece of armor. It is in the presentation to the personnel USING the protected vehicles/bunkers that needs to be a lower value of "nn" (exactly what would have to be determined from battle experience). This value would give the armor USER some idea of what are his odds of staying alive to finish his mission when using various options as to tactics. A V50 value means that his chance of doing this is very poor if he is hit more than once and only 50/50 if he is hit even once; poor odds. In fact, knowing several of these protection level ranges against various enemy weapons would be a great asset to determining what plan to attempt with the best chance of success.
Nathan, That looks like wisdom -- but if Eve had known about labour pains there wouldn't have been any more humans. (Evolution has conveniently given us incredible hormones which make somen forget so, miraculously, most families have more than one child.) You're quite right that we only consider events so that we can continue our thinking on to the consequences -- and of course our possible actions, and reactions, and their consequences. And their consequences... But obviously there has to come a time when we act. If 1001 tanks go into battle against 1000, you probably have almost 2001 crews who expect to be among the roughly 1000 victors. In reality, the morning after will find a few dozen of the winners and a one digit or zero number of the losers. Montgomery started the Battle of Alamein at about two in the morning and went to bed. An hour or so later he was woken up by one of his commanders reporting his horror at the carnage. Montgomery said "Yes. Carry on," and went back to bed.
@@TheDavidlloydjones I remember an article in the most reputed "Popular Mechanics" magazine (nod). An american pilot had just stepped out from the first tests of a Mig 29 and he was asked if the Mig was "on similar foots" with the F16 or... whatever. Answer was simple and Antifragile. "When you go up, you dont want to be even, you want to be The Biggest Gorilla in the Sky". Fat Tony statement, if there is one. (edited for typos)
Does anyone know what distribution is Taleb saying at 39:48?
26:15 While the logic of the contrast is reasonable, the example used leaves something to be desired; ebola is spread only through direct contact with bodily fluids and cannot be transmitted over the air, meaning that its risk could in theory be managed by a social mechanism that works similarly as "not sleeping with Kim Kardashian". In my view, a more powerful fat-tail example would be something that has never killed many people, but has the potential to do so by causing an unstoppable chain reaction once unleashed. For example, fewer people have ever died from nuclear weapons than have died in the seasonal flu each year, yet an accidental launch of one nuclear missile in modern times could lead to the death of a large portion of humans in a short period.
Is there a paper or article that contains and describes the examples given in the talk? I would love to read it
Yes, this is first chapter of Consequences of fat tailes
Just read all his books.
Love taleb. Taleb protect us , help us guide us amen.
not to piss on your bonfire but i was expecting smarter comments under an NNT youtube video, and yes i am fun at parties
24:04 " 'Pinkerism' is when you write 800 pages"
Link to his book that has all the details: www.amazon.com/dp/1544508050/ref=cm_sw_r_tw_dp_U_x_3PIGEb0XVZ6ZE PDF version: www.researchers.one/article/2020-01-21
When was the talk recorded?
Post Internet world.
Guys an idea how I start from scratch studying his concepts from a mathematical and pratictical approach? I read his books but it's clearly not enough to master the profond concepts.. Anyone can help? Thanks
I'm thinking of the same.
Hi dude it's the dumb Guy Jeff here with 9000 hard core evidence have to nationwide detective
Well if you really want to learn something about the mathematics of what he is talking about, you can read good (modern!) statistical physics/econo-physics textbooks as a basis, while always keeping his main points in mind. I would also recommend reading something on the extreme value theory, which is basically the framework to deal with the distributions that he talks about. This is closely related to large deviation theory, which I know a bit more about. I have got to warn you though, these areas are fascinating, but very dry when you are just starting out. I think most of the principles that he talks about can be followed without writing down a single equation.
Start trading options
Wow
58:53 "[Obama & Cass R. Sunstein]... were bothering Cameron, you know, the much [sic] group" What is Taleb referring to here? I can't recall.
I simplified a search by omitting 'Obama' and I learned of the Nudge Theory book Sunstein wrote in or around 2008, which Cameron borrowed from a decade later. Still doesn't explain the 'bothering' but it's a start, anyone remember their disagreements or criticisms? www.newstatesman.com/politics/economy/2019/05/cass-sunstein-and-rise-and-fall-nudge-theory
@@blackjackreward4456 Also Cameron is a friend of Niall Ferguson, who has talked about antifragility, specifically the concept of a network as opposed to a hierarchy. These ideas originate with Mr Taleb of course, post financial crash.
A gentleman with captivating insights who is simultaneously a complete tosser. He is akin to a brilliant theologian who reveals stunning connections, hitherto unknown, between biblical passages but who simultaneously is blind to the credibility of the source from which he draws all his inspiration.
An odd comparison. He is advising skepticism of knowable cause and effect but not recommending faith. That is almost the opposite of theology.
@@tavom6710 To draw a concrete conclusion from an analogy is to misunderstand the use of an analogy. My point has nothing to do with theology per se. One of many examples where he is a "theologian" should suffice to illustrate the general principle: in 'Antifragile' his many discussions ABOUT medicine are absolutely valid (these are analogous to the 'theologians connections between biblical passages') but he is utterly clueless about medicine itself, that is, he adopts an uncritical and unreflective attitude towards it (this is the 'Bible' for him). In his view we should apply antifragility concepts TO medicine but not question established medicine ITSELF. He talks about suckers in his books. He should add himself to the list.
"To draw a concrete conclusion from an analogy is to misunderstand the use of an analogy." Ancient Greece called and they want their catchphrases back. There have actually been other philosophers since Plato. “utterly clueless about medicine itself” medics are often clueless about medicine itself. Poincare didn’t know much about the weather but his mathematics disabused many climatologist of their hope of predicting it. Your message is a list of non-sequequitors ending with your initial opinion. Antifragile is the only one of the Certo I haven’t read. As far as I can tell you didn’t enjoy but haven’t read any of the others. I will take your remarks as literary criticism rather than philosophical analysis. Thank you for your input. The Black Swan is very good.
9:50 example
If the paper refers to moments, then shred the paper, and then it goes back to nature. Does he mean nature or Nature?
I don't think Taleb would make the distinction, but since you do - Nature would be my guess.
There's NO shredder in Internet world.
@@k.h.6991 The bloke is way out of date and not worth listening too.
What he says is that buy put options to benefit from black swan. I'm saying that sell put options SMALL . Payoff function will be in favour of you based on historical data. He says CEO is fragile for blackswan but taxi driver is not. I'm saying you better be CEO until blackswan. Payoff function will never put you below taxi driver income.
was this the college history department?
It was the department of the fittest.
45:00 oya
"So many americans die from smoking too much alcohol" 25:30
And most people die today from setting people up for crimes like me Jeff Vanessa future husband and feel sorry for 5 million of you I lived Italian life in NYC and the old Italian wise guys already know the 2000 top people oin charge and you deal with 5hem and criminal charged for fraud and call bkorg need trump has everything. Good luck Ahmed josh big 5 weak ass
30:00 profound
Introduce yourself first.
He is very unappreciated. His logic is sound, make government, companies, and civilization more resistant to catastrophes and "black swan" events.
He should drop the KK joke. But important lecturer
8:53
Taleb is the antithesis of lazy thinking.
Baws
I think that his problem is that he does not explain clearly. The flow is cut, a lot of digressions, the sentences are sometimes cut, jumps from slide to a digression then waves the pointer - like from 38:00. Difficult to follow. Having said that I think he is right but few people will make an effort to understand. Problem with presentation skills.
the camera work is giving me motion sickness.
Its as if you can shred paper in the Internet. I bet he wouldn't even know what Facebook was.
Thier maybe irony of the black swan, if the genetics of birds came from Gonwanland then filled the world. The black swan was first, a random events produced a white swan.
It just shows Europeans have no clue about 1/2 the reality. They're So busy dealing with history and tradition they're badly effected by that lifestyle. Unable to see what's arrived and tooo slow to adapt. If you listen you'll go down too.
"Donald Rumsfeld's `unknown unknowns'" is actually "the Connecticut Avenue greasy spoon fortune cookie."
The woman who appears at the beginning of the lecture is trying her best to sound smart.
😄😄
She’s smarter than you!
I've followed as much of Nassim's work as I can find. It's all brilliant - except in one area. At least as far as I can surmise. And I don't blame him for this, as I suspect he simply has not taken the time to think some aspects of his bias all the way through. For example, he endorses UBI. Oddly, this really stands in juxtaposition to his overall credo of "skin in the game." UBI, ultimately, is not...can never be, empowering. Rather, ultimately it is indenturing. Odd that he doesn't "see" this. But I'll give him a pass here on the one minutia as the bulk of his work is so immense and is liberating beyond that of countless others. Even with its freckles, I'll take the whole package any day rather than the dead-end toxic dogma of literally everyone else. Thx Mr. Taleb.
When you talk in riddles and codes all day the Internet is going to expose all your faults. You are what you do. Those secrets all leave deep scares on you. He's already hurt his own brain with all those tasks. He's dropping words and slurring others, etc. He's going to have to talk 10 times faster cause we've already consumed all his videos. He's got no time for this, cause this consumes all his time. We're only just beginning and already he can't handle Internet pace. Mentally, he's already broken. You don't have the option to stop for a drink and a sit down in Internet world.
I would pay big money to watch him try to explain the first 5 minutes to Trump lol.
Trump knows him
@@alfredthepatientxcvi he said explain .
10 minutes in, I have no idea what this man is talking about. Perhaps I am an idiot. There is no introductory preamble or context-setting for intelligent lay-people. As a public lecture, it would be helpful if the speaker did not assume everyone would know and already understand his area of expertise.
I don't think you are alone. The comments are very impressed, but fail to mention any concrete learnings. It's just a collection of philosophical-probability insights that are not actionable imo
he wraps up at 57 minutes... he doesn't like politicians, he thinks grandmas know more than professors, everybody is wrong, the end
I'm not sure this is a great help for anybody.
This guy is a "low talker". I can barely understand/hear what he is saying.
He’s said this work to his advantage as people have to listen harder to understand him. Which makes people switch into active listening mode from passive listening.
Yeah! This talk is very irrelevant.
Fraud and repulsive