- ‹ previous
- 1153 of 2794
- next ›
The Black Swan
I finished The Black Swan over the weekend, and although a full review at this late date is kind of pointless, it's an odd enough book to deserve a few notes.
First, the tone: it's intensely annoying. The problem is that Nassim Nicholas Taleb basically sounds like a crank. His prose has all the usual markers: everyone else is an idiot (this includes philosophers, economists, historians, journalists, and pretty much all social scientists, among others); he's the only one who truly understands the world as it is; there's a monocausal explanation for this almost universal lack of understanding in others; and there's a tiny cast of other unappreciated geniuses who do get it (Benoit Mandelbrot, Karl Popper, G.L.S. Shackle, Daniel Kahneman, etc.). It's all sort of Unabomber-like, though with a better sense of humor.
But of course, that's just an esthetic judgment. What about the content? Well, here's the funny thing: once I got past the tone I didn't really have a problem with most of it. Taleb talks about confirmation bias and the narrative fallacy. Survivor bias and the anthropic principle. He writes about how humans are hardwired to be bad at estimating risks in the modern world. He explains how network effects can create large inequalities out of small differences and how randomness is responsible for more of our success than we think. As it happens, this is all stuff I was already pretty familiar with, which made the book annoying and a bit tedious, but that's obviously not Taleb's fault.
Unfortunately, I'm not sure how effective the book would be even for someone who found this stuff new and interesting. Taleb tends to flit from subject to subject without ever really explaining anything fully enough to make sense, and in the end it's not quite clear what case he wants to make. Generally speaking, he wants to persuade us that we know less than we think and that forecasting the future is a mug's game because history is primarily governed by huge, unpredictable events that come out of nowhere (black swans). But this is sort of a banal point: scholars have been arguing about the importance of contingent events vs. broad historical trends forever, and the difficulty of predicting technological breakthroughs is well-trod ground. Worse, Taleb doesn't add much to what's already been said about it. Just the opposite, in fact. In one chapter he cherry picks some inventions here and there to help make his case, but even using his own hand-picked examples he's not very convincing. We all know that penicillin was discovered by accident, but the computer? Taleb seems to think it sprang out of nowhere, but that's sure not how I remember it. It was a big invention and a huge discontinuity, but it was hardly unpredictable and hardly an accident.
The last few chapters are a diatribe against statisticians who are over-devoted to Gaussian distributions, and I don't have the chops to know if he has a point there. The statisticians I've come across all seem to be keenly aware that there are lots of different kinds of distributions in the world, but maybe that's all shuck and jive. Maybe they talk a good game and then end up modeling the world using bell curves anyway. And in the financial world, for which he reserves his primary scorn, I know even less. Taleb says they obstinately continue to use Gaussian models that flatly don't work and hide known risks, and he probably has a good point. Certainly recent events are on his side, and Wall Street's almost cultlike reliance on VaR and CAPM and portfolio theory and the Gaussian copula seems to have played a big role in its current collapse. But who knows? Maybe it was actually just a gigantic housing bubble and this other stuff is a minor sideshow. I don't know for sure, and Taleb doesn't provide enough evidence one way or the other for me to make up my mind.
In the end, he doesn't have much advice for us. He insists that Mandelbrot provides a better mathematical basis for financial modeling, but never tells us how. He insists that the world is mostly governed by a small number of big events, but never seriously grapples with the arguments for or against. He insists that he's used his sensitivity toward black swan events as a practical guide to his own trading and investing strategy, but then he sums it up with this: "As I said, if my portfolio is exposed to a market crash, the odds of which I can't compute, all I have to do is buy insurance...." Really? That didn't work out so well recently, did it?
(On the other hand, the rest of that sentence reads: "....or get out and invest the amounts I am not willing to ever lose in less risky securities." That's basically a way of saying that investors should limit their leverage, and he's certainly right about that.)
So I'm not sure what to think. Taleb makes plenty of good, if rambling, points about the limitations of human nature, but his concrete advice is pretty prosaic. Stay open to lots of experiences. Embrace empiricism and let the data lead you without pretending it says things it doesn't. Keep the possibility of massive losses in mind and don't invest money you can't afford to lose. If you are going to take risks, invest in things like startup companies, where the risks are plain and open. I don't really have any argument with most of that, but I'm not sure any of it is really all that remarkable either.
Did I miss the point? Maybe. To be honest, I'm really not sure.
well
Well, if I were to go on your review, and some of the other comments I've seen in these Black Swan threads, I'm certainly glad I chose not to read any of his books. All I've read from him are shorter articles (finance focused mostly) and op/eds, and those seemed pretty solid (at least well ahead of standard analysis, which seems still lost in linear projection), and perhaps there's just not enough there to really fill a 200 or 300 page book, at least without musing, asides, repeating, and filler.
For those interested in more non-linear, complexity-oriented economic analysis, I highly recommend this book, I personally found it very enjoyable and informative (even though I was familiar with most of the chaos/complexity elements already), at least from a layman's perspective (I'm sure there are other economics-related complexity volumes that are more thorough or recent).
Butterfly Economics
About Mr. Mandelbrot
tagged as:- solution
Kevin Drum >"...He insists that Mandelbrot provides a better mathematical basis for financial modeling, but never tells us how...."
Mr. Madelbrot tells everyone in his book "The (Mis)behavior of Markets: A Fractal View of Financial Turbulence" which is a reasonable read w/plenty of references for those that really want to "dig in" and follow his evidentiary trail.
Give it a try even if you don't "get" math.
"Fear not the path of truth for the lack of people walking on it." - Robert F. Kennedy
I think he should rail
I think he should rail generally against the charlatans, the physicists and applied mathematicians who (ab)used statistics to construct all these bogus instruments that became lethal weapons in the hands of the Harvard MBAs, quite analogously to Yoo et. al's perverse provision of legal cover for war crimes.
Bringing Gauss and Student into it is absurd.
You just wrote his book.
In eight or so paragraphs, you pretty much covered what it took Taleb a book to lay out. And in better prose.
further reading...
Mr. Madelbrot tells everyone in his book "The (Mis)behavior of Markets: A Fractal View of Financial Turbulence" which is a reasonable read w/plenty of references for those that really want to "dig in" and follow his evidentiary trail.
Great point.
Sometimes it seems we get too caught up in wanting one book to explain as much as it possibly can, when it's actually unfeasible for that to occur in any book, let alone a layman's book.
In my own experience, when I read short articles or op/eds from Taleb, it was informative and enlightening for me, because I basically took the finance-related stuff he was talking about (which I was least familiar with) and followed up with my own research (through Google, other books, etc.).
And, since Taleb is throwing out all the latest jargon and what not, he certainly improves the efficiency of following up the arguments he presents in more depth through further investigation (and keyword searches).
Even penicillin wasn't
Even penicillin wasn't really all that random. In 1929 or so (going from memory here from a book read years ago), Alexander Fleming noticed that a penicillium fungus had interesting properties that suppressed some bacteria, wrote it up briefly, and forgot about it. (Later literature searches showed that others had made similar observations before, and even applied it medically). In 1940 or so, Florey and Chain went looking for something with good antibiotic properties, pulled a culture of Fleming's strain out of the lab storeroom as one possibility, and and did some good productive science. The Fleming "miracle" was then retroactively overhyped by skilled PR types. The reality was more like normal plodding white-swan science at work.
Mandlebrot
Kevin is no mathematical innocent. He did manage to get into Caltech, even if he didn't stay for a degree.
Gauss and the Central Limit Theorem
So what does Taleb have to say about the Central Limit Theorem (CLT)?
If the number of samples of a random variable is "large," the CLT proves that the Gaussian distribution makes a good approximation of any distribution.
So when Taleb finds fault with statisticians for overusing the Guassian, is he talking about small 'n' statistics, or is he finding some other problem? Or is he confused?
Just curious. Not going to read the book.
* Large, in this sense, can be as low as 30.
Completely agree, Kevin!
In fact, as one commenter points out, you summarized the book without the irritating tone. Frankly, I couldn't read the damn thing beyond 50 pages.
Taleb wasts your time
I think you summarized it pretty well. It's not that Taleb is stupid, it's just that he is incredibly full of himself and contributes pretty close to nothing original himself. I think of him as a popularizer of other people's work, usually without giving them proper credit.
As a finance guy, let me add that his entire spiel about how everyone in finance blindly believes in normally distributed asset prices is utter and complete nonsense. I got my finance PhD a long time ago, and even then you'd learn about fat tails, stochastic volatility, stochastic correlations and all that jazz in the very first semester of the program. Heck, we are teaching it to our MBA students in the intro investment class these days, and have for probably 15 years. Oh, and we have been teaching VAR as an example of how not to do risk management for about as long.
If you don't believe me, check out "An Overview of Value at Risk" by Duffie and Pan in the 1997 (!) Journal of Derivatives. It's a review article, which means it doesn't do any original research and just summarizes the prior literature. It's a long time that I read it, but as far as I recall about 80% of the article is about how VAR fares in a world with non-normal distributions and unstable correlations. In other words, we really didn't need Taleb to tell us about this, nor would you ever learn from reading Taleb that all the correct parts of his spiel are old hat.
Jimm: I agree that Taleb
Jimm: I agree that Taleb isn't obligated to explain Mandelbrot in detail. The problem is that he doesn't really explain Mandelbrot at all. You'd have to read the book to see if I'm being fair about this, but he says virtually nothing about Mandelbrot aside from (a) a few paragraphs about self-similarity, and (b) a bit about how extreme results have higher probabilities than under normal distributions, and (c) how he was ignored by the establishment in the early 60s. It's not even remotely enough for a reader to get a sense of what his math is about and how it might be applied to financial problems. Even for a lay audience, he needs to present at least that much.
Teece: His main issue is that he thinks Gaussian distributions are used in areas where the distribution isn't really Gaussian. He doesn't have a problem with modeling, say, height as a normal distribution, but he *does* have a problem with modeling financial performance as a normal distribution. It's not a matter of the number of data points. It's large in both cases.
!!!
teece,
The Central Limit Theorem only applies if you have a distribution with finite variance. This makes a really really big difference!!
For example, if you have a series of independent standard Cauchy distributed variables, the sample mean will in fact be Standard Cauchy distributed, no matter how large your n. There will never be any convergence to the mean!
This was the gist of Mandelbrot's work really. He showed that Cotton future prices were Alpha-Levy distributed with alpha=1.7, a distribution with infinite variance. This meant that using the central limit theorem in work would be very very wrong.
Mandelbrot's point, and Taleb's also, is that distributions with infinite variance, where the central limit theorem doesn't apply, are very very common, particularly in finance, and that this changes everything, making it very important to know precisely what distributions we're dealing with.
However, and Taleb has written about this in academic papers extensively, the "inverse problem" of identifying distributions from data is very hard. And even when we can do that, estimating the parameters of such distributions can be difficult or even impossible.
Of course, this isn't really news to anyone in the field. But it poses a rather difficult problem, and solutions are controversial...
Bistro Distro, let's call the whole thing off.
FT is running excerpts from Ms Tett's book "Fool's Gold." The latest explores some of these issues:
As soon as Duhon talked to the quantitative analysts, she encountered a problem. When JP Morgan had offered the first Bistro deals [ur-CDOs] in late 1997, it had access to extensive data about all the loans it had pooled together. So did the investors who bought the resulting credit derivatives, since the bank had deliberately named all of the 307 companies whose loans were included. In addition, many of these companies had been in business for decades, so extensive data were available on how they had performed over many business cycles. That gave JP Morgan’s statisticians, and investors, great confidence in predicting the likelihood of defaults. But the mortgage world was very different. For one thing, when banks sold bundles of mortgage loans to outside investors, they almost never revealed the names and credit histories of the individual borrowers. Worse, when Duhon went looking for data to track mortgage defaults over several business cycles, she discovered it was in short supply.
While America’s corporate world had suffered several booms and recessions in the later 20th century, the housing market had followed a steady path of growth. Some spe cific regions had suffered downturns: prices in Texas, for example, fell during the Savings and Loans debacle of the late 1980s. But since the second world war, there had never been a nationwide house price slump. The last time house prices had fallen significantly en masse, in fact, was way back in the 1930s, during the Great Depression.
The lack of data made Duhon nervous. When bankers assembled models to predict defaults, they wanted data on what normally happened in both booms and busts. Without that, it was impossible to know whether defaults tended to be correlated or not, in what circumstances they were isolated to particular urban centres or regions, and when they might go national. Duhon could see no way to obtain such information for mortgages. That meant she would either have to rely on data from just one region and extrapolate it across the US, or make even more assumptions than normal about how defaults were correlated
JPM decided that these were too unpredictable, and stayed away after that (almost).
"Generally speaking, he
"Generally speaking, he wants to persuade us that we know less than we think and that forecasting the future is a mug's game because history is primarily governed by huge, unpredictable events that come out of nowhere (black swans). But this is sort of a banal point"
But this is NOT a banal point!
If you aren't sure about what the future holds, you build systems that have plenty of slack, plenty of capacity to absorb shocks. It's only when you are convinced that you know what is ahead of you that you whittle down your shock absorbing capacity to nothing.
There are circumstances where this makes sense --- fluid mechanics hasn't changed much in the last forty years, and is unlikely to change much in the next 40, so it makes sense to successively whittle away the excess aluminum in a soda can, giving us today's marvels which are what, a half, a third of the weight of their predecessors.
There are other circumstances where this makes much less sense ---- finance being an obvious example. And yet large numbers of apparently very smart people convinced themselves that they did, in fact, understand the nature of change in the economic and political world sufficiently that they could bet trillions of dollars on what would happen next.
And it's not just finance. The entire Bush administration was about believing that they knew, with absolute certainty, what would happen if they tore up 70+ years or American experience and replaced those ways of doing things with drastically different alternatives. The entire project was predicated on simply ignoring possibilities like "maybe Iraqis won't love us", "maybe global climate change is real", "maybe peak oil is real", "maybe there are some things government does better than the private sector".
True enough
but probably 99 percent of the rest of the world knows those things were crap. IOW, he's writing for an audience that doens't need to be convinced. Does he think the ex-adminstation and former finance "geniuses" are going to read the book and slap themsevles on the forehead?
Political aspect of black swans
People in the finance world are not stupid and yet somehow buy into the current system of economics, whatever it may be on a given day. Then, after Phil Gramm et al change the system strange things start happening and everybody yells 'black swan'. No, not really.
So, how likely is it the public would be different and not fall for the same kind of intellectual mistake when the Republicans, led by the Bush family, 'create reality', change the system in violent ways and lead people to believe it's a naturally occurring strange black swan and they are the saviors? How is it possible the Republicans could yell "cut taxes & small government" for decades and fully support the Bush (and Reagan) deficits & mounting federal government debt?
Manipulating people is NOT leadership. Lying to people is rarely good for those people. Blindly following or believing anyone who claims to be a leader can be very bad for your health and welfare.
Black swans might be a popular idea, but in reality -- not the Bush reality -- must by definition be quite rare. To quote Condi Rice, nobody can have predicted it if it's really a black swan. Of course, all too often, during the Bush era the unpredicted was very carefully created beforehand and then very well exploited as it unfolded.
Sorry, forgot the <p> tags...
teece,
The Central Limit Theorem only applies if you have a distribution with finite variance. This makes a really really big difference!!
For example, if you have a series of independent standard Cauchy distributed variables, the sample mean will in fact be Standard Cauchy distributed, no matter how large your n. There will never be any convergence to the mean!
This was the gist of Mandelbrot's work really. He showed that Cotton future prices were Alpha-Levy distributed with alpha=1.7, a distribution with infinite variance. This meant that using the central limit theorem in work would be very very wrong.
Mandelbrot's point, and Taleb's also, is that distributions with infinite variance, where the central limit theorem doesn't apply, are very very common, particularly in finance, and that this changes a lot, making it very important to know precisely what distributions we're dealing with.
However, and Taleb has written about this in academic papers extensively, the "inverse problem" of identifying distributions from data is very hard. And even when we can do that, estimating the parameters of such distributions can be difficult or even impossible.
Of course, this isn't really news to anyone in the field. But it poses a rather difficult problem, and solutions are controversial...
Fair enough Kevin
I sorta overemphasized the "not needing to explain everything" part, when I was really more attempting to follow a thought that it's a much more practical use of time to read shorter articles/tomes and follow them where they lead.
In that sense, we couldn't be in more agreement about this book, or in our habits as far as reading physical books, though I would be more interested in my Kindle having uninhibited, high-speed access to the Internet, so I could jump around following various trains of thought engendered by the reading (at least for non-fiction, I don't really read fiction much anymore).
With that in mind, what kind of internet access do you have on Kindle? I don't have one, so have limited knowledge. I know it obviously has some kind of connection to to get the books, and have heard something about access to Wikipedia, but I would need complete access to the Internet as far as my reading habits go, or at least access to a great set of encyclopedias, scientific journals, and things like PubMed.
Power laws vs. Gaussian
The difference between power law distributions and Gaussian distributions is pretty damn significant. An event that you think to be 10 sigmas unlikely might only be four or five. That's huge. And the Wall Street "rocket scientists" definitely care about this kind of thing.
However, Gaussian distributions are far more well known and well developed, and consequently easier to calculate with, and as long as stuff deep in the tail doesnt really matter very much, there's no reason to redo it with power law distributions, assuming you even understand the differences.
Which is not to say that the subject was developed well in The Black Swan.
The model of the world that
The model of the world that would best explain the recent meltdown (as well as the inevitable future ones) is probably not going to be one that analyzes the deep structure of the world's economy or of the markets. Rather, it would be a model of the probabilities of financial success (and avoidance of penalties) for those market participants who make enormous bets using Other People's Money.
If you look at any one of the major players at places like Citi, AIG, etc., you will find them still enormously wealthy today, with slightly bruised egos (though they all have some explanation) but not much poorer for it. Future players, if rational, will follow the same path unless the risk they take on behalf of others somehow becomes personal, retroactive risk -- which is unlikely to happen.
Perhaps a more mathematical way to put this would be to say that, given a particular model used as the basis of investment/trading decisions, the distribution of outcomes will be very different for those making the decisions than for their clients and and for society as a whole.
Re: Other People's Money
To suggest Taleb predicted the crash of 2008 is ludicrous. He's talking about Big Things Happening Out Of The Blue. But the economic crisis wasn't a black swan. It was caused by people investing recklessly because they were using other people's money, as JS put it, and took on no risk personally. There was no penalty for failure for those who fooled with investors' money while "earning" massive margins.
I haven't read the book. But for those who have - does Taleb talk about the need for greater regulations in the finance industry? That would certainly cut down on the extreme variability and make finance boring again, as Paul Krugman puts it.
I reviewed "The Black Swan"...
...for a magazine pre-publication and could not get past Taleb's insufferable tone -- the supercilious contempt for “academic philistines” and “philistine scholars” with “standardized minds” who ask “ludicrous” questions, who are “middlebrow” and “dogma-prone.” We are to pity him, a wise man condemned to spend his life surrounded by fools -- us, that is. “Every morning the world appears to me more random than it did the day before, and humans seem to be even more fooled by it than they were the previous day," he writes. "It is becoming unbearable. I find writing these lines painful; I find the world revolting.”
And Taleb’s intellectual preening—his stated (boasted) facility with languages, high concepts, statistics, logic, philosophy, and high finance—is intolerable.
Plus, I did find his thesis banal and obvious, with many of his examples flawed or flat-out inaccurate. (As examples of black swans he cites “the computer, the Internet, and the laser”—all “unplanned, unpredicted, and unappreciated upon their discovery" . . . except that they were hardly unplanned—people spent years conceiving, developing and applying them.) His argument really can be boiled down to saying that random things happen, and that we should resist creating narratives that don't allow for randomness. And Taleb takes this way too far: He seems incredulous that anyone even tries to explain unusual occurrences and shifts, but that’s what historians, for instance, do.
I knew people would at least pick up the book, since Taleb's debut was a left-field hit, but remain surprised that almost anyone -- anyone not being paid to do it, that is -- has been able to spend four hundred pages in his company.
Oh he speaks the languages
No need to engage in petty comments on that point, but his generralised rage is quite unhelpful.
Miroring Wall Street habits.
missing the point Kevin
Sure his writing is not so good, but you are missing the point. His book was published in 2007 BEFORE the stock market/subprime/ financial meltdown. He predicted the level of current market distress, at that time. I recall it very clearly because I had just read his book, and consulted a Wall Street-working MBA friend of mine. He said the word on the street was "economy is great, corporations are flush with cash, no problems on the horizon."
Of course his comments seem banal in May of 2009, AFTER everything he pithily predicted has occurred. Give him his props for his prescience, if not for his writing and personal qualities.
@Commenterlein: great, finance types know all about this stuff. So you say. Why then are we in the problems we are in, given that all the bankers quant types have been using flawed gaussian distributions to model the risks? Yes, a lot of Germans knew Hitler was a lunatic as well, but it didn't seem to help much- most went a long with him in any case.
It seems to me that quantitative finance does have to shoulder the blame. Sure, they might teach power laws, but if they are not used by anyone, what's the point? Excuses seem a bit ridiculous at this point.
No, I think you're missing
No, I think you're missing the point here. Taleb's reasoning is banal because whether or not his predictions were accurate, there's nothing new or interesting in his reasoning.
Worse, there's nothing USEFUL in his book. Except to the extent he refers you to other, more cogent, thinkers, Taleb's point is essentially "Complex things sometimes break for unpredictable reasons." Banal. Tell me the last time the first piece of financial advice you heard was "Diversify," and the second was "Make sure you have some downside coverage." Seriously, his points are stuff I learned in a High School economics couse. HIGH SCHOOL.
It gives us common vocabulary.
I read Talib a few months back. It certainly was irritating, and coming from a physics background where most distributions are gaussian like (or maxwellian), or sommething else that can be derived from first principles especially so. Maybe in economics/finance his claim about people using bogus distributions simply so they can make progress, and sell themselves to hedge funds is true, but I think of finance as just a small corner of the world containg unusually greedy people.
But, I did feel it was semi-required reading. Simply because it gives us a common set of examples and vocabulary with which to discuss such issues. Otherwise, I think most people who have taken the time and effort to think through the issues would have already reached similar conclusions.
If they cite Mandelbrot or
If they cite Mandelbrot or the Heisenberg uncertainty principle, it's just gotta impress you.
Maybe
I read the book at a good time, last summer, nothing to do but kill some time in the Spanish sun. Not coming from a mathematics background I found that it was written to my level, understandably if not poetically. Then again, I don't seem to have a problem with his style of over-the-top finger pointing at the idiots; thinking back now, it would be too much for many. He seemed to hit the nail on the head though with quite a few points as the financial world's unraveling was picking up pace, even inspired me to write a post about it!
In Case You Missed It
I agree with Commenterlein's
I agree with Commenterlein's comment above. Taleb is pretty close to worthless. He also lost a ton of money and sunk his original hedge fund in 2006, although he rarely mentions this when talking about his 100% return last October (big whoop)
I've only seen interviews
with him, but my impression jibes with your book review. I found his points obvious, his arguments poorly made and the core concepts derivative. I had thought maybe the book was really great and just didn't translate well to short interview format, but I guess not. Which leaves me wondering why he's taken so seriously and everyone talks about his book like its some sort of visionary work. Nothing Kevin mentioned seems like an original idea at all. Its more like he read a bunch of articles in New Scientist and The Economist and summarized them.
I've only seen interviews
with him, but my impression jibes with your book review. I found his points obvious, his arguments poorly made and the core concepts derivative. I had thought maybe the book was really great and just didn't translate well to short interview format, but I guess not. Which leaves me wondering why he's taken so seriously and everyone talks about his book like its some sort of visionary work. Nothing Kevin mentioned seems like an original idea at all. Its more like he read a bunch of articles in New Scientist and The Economist and summarized them.
Gauss and the Central Limit Theorem
Thanks a lot for that, David. I knew the CLT had conditions, but in the stat. work I had done, they never were violated -- so I forgot what they were!
I appreciate the clarification.
Taleb makes most sense read in the context of Risk Mgmt
As practiced on Wall Street, not statistical theory as such. His original book was far more useful, and I am afraid his arrogance led him to probably reject helpful editorial improvements to his ranting.
He does, however, paint with a baod bush.
That being said, although yes people are aware of the limits of many statistical tools, and over-reliance on things like VaR for applications it was not really meant for, the reality is that under pressure, people "work around" or "assume" good approximation. In addition, regulators have writtern in some approaches, so there is a small aspect of truth to the otherwise generally lunatic accusation by the American libertarians that Gov't "caused" this. Certainly regulation, intended to limit risk formalised some approaches that have proven to have deadly limits.
On computers . . .
Did he really cite an example that computers came out of the blue? Even for PCs this is simply untrue.
Computing history is something I happen to know about, and the idea that any computing devices sprang fully formed, Athena-like from the head of Zeus is totally false.
Please tell me he didn't use the Apple II for his example. Please.
Tripp
shuck and jive
That's funny.
Use the bibliography of the Black Swan
I seem to be in the minority here and thought the book was really useful. But then, I read it three years ago. His points may seem banal but I think that's more because they are very simple yet unintuitive so calling something banal allows you to dismiss it out of hand. People really do not want to hear what he's saying as it runs counter to the way we are programmed. We WANT to believe in stories, we want everything to make sense but that does not make it so. You can find the narrative fallacy and confirmation bias and all these issues elsewhere (and written in a less abrasive way) but please find me an instant where his general ideas are wrong. At the very least, check the bibliography as he makes some excellent recommendations, including Karl Popper's "The Open Society and Its Enemies" (I'd say required reading for readers of this blog) and Daniel Kahneman's work.
Correct
A lot of Taleb's writing is really insider talk that makes real sense (in terms of his banging on about how no one is listening, aware of the issues) only in the contet of risk management and modelling practice "in the wild."
I don't think points are banal
I just think he's taken very important points and ideas that have been presented much more clearly in other places, thrown them together (very poorly according to most posters here) and then got himself a nice book deal. And this is fine with me. But what I don't understand is why he garners so much praise and recognition for doing this. Yes, his book came before the financial collapse, but I really don't think that's such a great feat either. I think many people saw the underlying problems but chose not to speak out about it because they were getting rich, or did speak out and were ignored.
Finance versus Civil Engineering
It is difficult to refrain from comparing Finance to Civil Engineering.
From the sidelines what I can glean is that it is as if there were a global standard for bridge building, one which had stood the test of time, but allowed modest profits. Then a new standard was proposed, a new standard which was 'modern' and 'better' and 'ingenious,' a standard which quickly became accepted and globally adopted. In addition to building new bridges under the new standard, many existing bridges were replaced with new bridges using the new standard.
Why the quick adoption of the new standard? Because it allowed the bridge makers and the bridge designers and the toll takers to become very wealthy very quickly. With wealth came praise and fame and they were lauded greatly.
Then the new bridges began to fail, all around the world, in rapid succession, with great cost to the public.
And now the bridge standards makers are saying "only we can fix this, you need us," and saying "don't even think of going back to the very old, tried and true way. We need supplies and money so we can get back to the newer way so we can start making all that money again."
If laymen complain they say "You are ignorant. You don't understand why this happened. Trust us, we know what we did wrong. Trust us, or you'll have no bridges at all. Give us your money and start us up again or transportation will cease and you'll all pay the price. Is that what you want? You should be grateful to us. Without bridges civilization will fall."
Tripp
Financial Engineering is not Engineering.
The comparisons don't really make sense insofar as the underlying basis is fundamentally different.
The fundamental difference is finance is in the end about human interaction, not environment and materials interaction. That is what wrong footed modelling and what makes the comparison with engineering standards at best inexact if not entirely false.
Unlike steel, for example, which has known properties, human behaviour is changeable, the modelling on mortgages for example assumed a number of behaviour patterns that large and deep data sets proved out as long-standing. And they were, until they changed.
Further, as to your item on laymen and critiques, much of the problem is that laymen are often attacking the wrong issues and lack an understanding of where deeper issues lie. That is natural given the subject matter, but could end up producing fundamentally non-helpful actions, regulations.
I would point out that Junk Bonds produced much heated and angry commentary on the part of the lay population, in particular driven by journalists highly incomplete and sensationalistic presentation of the then new product. Afterr being put in the right regulatory box, high risk bond financing to 'sub prime' companies has proven quite effective. Once brought into the right framework much of the hated innovation I suspect will prove quite useful.
Isn't a lot of this a
Isn't a lot of this a function of when you were reading it? Published in early 2007, it functioned as a lightning rod for a lot of people who were worried about the state of the financial markets. Maybe that function ended up giving it a bigger reputation than a straight reading post-crash gives.
Lounsbury, If, as you say,
Lounsbury,
If, as you say, models of human behavior are much less reliable than models of the physical universe, then why do we, as a society, allow finance to have so much power?
It seems to me if we have something we need such as commerce and we know that it is hard to model then why don't we put in extra safeguards? Why do we allow the financial people the ability to rock our world?
Tripp
Economists are interested in
Economists are interested in non-Gaussian distributions. Financial people are not. They are interested in getting a NUMBER. My wife, a PhD in Mathematics, discovered this. She collaborated first with some economists, and they were very interested in nonlinear non-Gaussian processes in pricing. Then she moved on to collaborate with some financial types. These people only wanted a result that was generated in some superficially plausible way. As quickly as possible, and with as little cost as possible. That was it. Whether the number was actually defensible was not even a matter for discussion. She lost that job when she tried to start that discussion.
thnks for your post. it's
thnks for your post. it's wonderful.....
http://www.bloomberg.com/apps
http://www.bloomberg.com/apps/news?pid=newsarchive&sid=aDVgqxiT9RSg
In light of the above link, this bit of facetiousness seems very poorly researched:
He insists that he's used his sensitivity toward black swan events as a practical guide to his own trading and investing strategy, but then he sums it up with this: "As I said, if my portfolio is exposed to a market crash, the odds of which I can't compute, all I have to do is buy insurance...." Really? That didn't work out so well recently, did it?
indir
thanks for motherjones
tiffany jewelry It seems to
tagged as:- solution
- result
tiffany jewelry
It seems to me if we have something we need such as commerce and we know that it is hard to model then why don't we put in extra safeguards? Why do we allow the financial people the ability to rock our world?
tiffany and co
Black Swan is a great book.
Black Swan is a great book. Thanks for your contribution. I wish more people can read it.
Want to upgrade your image and gain confidence quickly at affordable price?
Welcome to our site for replica watches. We have high quality replica watches on sale with 24-hour online customer service and free express shipment worldwide.
Franck Muller replica watches|
Girard Perregaux replica watches|
Glashutte Original replica watches|
Graham replica watches|
Gucci replica watches|
Guess replica watches|
Hamilton replica watches|
Hermes replica watches|
Hublot replica watches|
Thank you.