on 26 August 2011
This is an interesting read and likely worth it. The main point is there are a lot of people in finance who have no clue about statistics.
One of the key points is that many financial markets follow power law distributions. I was disappointed to see that no empirical evidence is presented to support this. So I tested it myself. 108 years data of equity returns in 5 countries (the DMS data set). There is no evidence to suggest annual equity returns follow a power law distribution. A much better fit is achieved with exponential.
Best to check oneself; free data is available from yahoo; R statistics program is free online; the fExtremes packages quickly tests if data follows a power law distribution.
Very disappointing and left me suspecting the other unsupported arguments in the book are similarly flawed.
on 10 March 2009
I greatly enjoyed the first chapters of this book: Mr. Taleb was presenting some challenging ideas to science, challenging some existing discourses and the supremacy of expertism and statistics of our time. For a social scientist like myself, it was a refreshing place to start on some challenging ideas that I think all serious social scientist graple with.
But as I finished the book, I saw that Mr. Taleb never got beyond expressing the idea, and that made it awfully superficial.
The second part of the book was outright annoying, as Mr. Taleb falls in the trap of being an "empirical sceptic" to everyone but himself (funny how scepticism does not seem to apply to his own ideas) and in the end provides absolutely nothing new to scientific epistemology but a "self-development" pseudo-pocket phisosophical explanation about how we all should live life waiting for the unexpected...
The ending is thus hugely disappointing for everyone looking for a serious challenge to the leading epistemology of our time, and Mr. Taleb's good humour and stories end up like a financial buble "Black Swan" that I was certainly not expecting at the start of the book.
on 13 December 2012
Quite a fascinating read and also quite a heavy read. The thoughts which Taleb has are very big thoughts which sometimes feel very sound, although you will probably find yourself shouting at the book, screaming `NO TALEB NO how can you say that' but I think that adds to the fun of reading this book, it is good to be challenged now and then and honestly I think that the basics of what Taleb says are all sound it's just that maybe the conclusions aren't so watertight. Still it is a book that demands multiple reads with a pen and paper to hand.
The main thread of this book looks at probabilities and how we toil under the impression that we can make good predictions, when in fact the variables are often too huge to ever be able to predict usefully. Taleb does make clear that he has no problem with predictions and statistics and the use of them, his problem stands with our blind faith and our inability to see that that faith constantly fails us.
The Black Swan represents the unknown unknown, it is the thing that we cannot predict and that we cannot plan for. Read the book and learn to be more humble.
Everybody knows this is a lazy sequel to a succesful book.
But I never read Fooled by Randomness, and I hold a couple math degrees, and I even once took a class from a hydrologist, so I reckon I'm well-equipped to pass judgement.
The main issue with the book is that if you read the inside of the jacket carefully you don't need to read the rest of the book. There's one idea here and it's that you never know if you've looked at enough data. If you presume you do, you run the risk that you'll get a surprise which upsets your world theory. That surprise is a bit like the first time you encounter a black swan. Turning that core idea into a (second!) book is indeed a stretch.
The second issue is that the author appears to be rather conceited / self centered and just keeps repeating how right he's got this idea and how wrong everybody else has got it. It gets tiring. He does not even get that point entirely right either. Lots of people's job description is to project current trends an hour forward. It ain't their fault.
The third issue is that it's a quote a minute from all sorts of sources and after a while you get the feeling the author is kind of trying to convince you he's covered all the bases. Or maybe that he's a truly erudite renaissance man who's done a lot of reading. Hey, maybe he had a word limit. You can't get there just by playing with the margins like we all once did in college.
On the plus side, the book is not as dreadful to read as the critics have it, and THE MAN IS RIGHT about the point (note that I'm not using the plural) he makes.
If you're stuck on a plane and the guy sitting next to you puts down his copy of the Black Swan, don't be afraid to pick it up, it's OK.
on 8 January 2009
Never before have I felt the urge to advise amazon buyers not to buy a book. But there's a first time for everything. Sadly, this book is an intellectual and stylistic mess.
In contrast to Mr Taleb's previous book (Fooled by Randomness) Fooled by Randomness: The Hidden Role of Chance in Life and in the Marketswhich I rate as a five star buy this book has nothing new and isn't even funny.
The basic argument is apparent from the title: just because you've never seen it before doesn't mean you won't see it tomorrow. Swans were assumed to be always white, until the discovery of black swans in Australia.
There are lessons to be drawn from this critique of what philosophers call the problem of induction, but not enough to fill a book. Instead, the author lets off a scattershot at several bats in a belfry.
First in the firing line is Plato. (Perhaps it's neo-Platonists, but an argument constructed on terms such as "What I call Platonicity" makes it hard to tell. The level of debate has already descended to the sophistication of a coconut shy.)
Second is the normal distribution. True, a roll of the die shows you one to six with equal probability. But two dice? Now we're talking a bell curve. The author seems to think that noone has wrestled before with the problem of extreme odds-against results, but's that's the future for you: hard to predict. A 95% confidence level is just that: once about every twenty times you're going to get an unexpected result. Surprise? Often. It is argued that these unusual unwelcome results are discounted from conventional thinking, and that the catastrophic consequence makes the tail of the distribution disproportionately important. Very true, but not new. Mountaineers and medical researchers (to name but two) have been confronting this problem for a century.
Not quite argued (but implied) is that all statistics is balderdash. Well, it's a branch of mathematics that deals with uncertainty, so we can't be sure.
Stylistically the book is all over the place. Reminiscence of the author's time as a "quant" (a statistical analyst of financial markets) elbow aside pseudo philosophical discussion, only to give place to some boasting about conferences he's addressed on the strength of his previous book. In between we get some noveletish stuff about "Fat Tony", "Yevgenia" (author of a bestseller that sounds unreadable), and a "fictional" character called Tulip. (Who is surely a direct descendant of Lupin, Mr Pooter's son in diary of a Nobody.)
We just can't predict, is the sum of all this. Indeed not, but we can bet that this book will be out of print just as soon as financial bestsellers go from "whoever would have thought it?" to next year's conspiracy theories.
The author thanks his editor, a certain Will Murphy. Mr Murphy, you didn't do your job, did you?
on 22 October 2007
NNT's use of a "narrative" strategy based on a questionable premise (did the sighting of the first black swan in Australia really impact the world?) to attract readers has paid dividends and his book has attained Grey Swan status among Amazon's Business & Investing bestsellers. The author, as he admits, has clearly enjoyed the writing experience which, besides dealing with the Impact of the Highly Improbable, has enabled him to sideswipe individuals and groups he sees mired in a world of Gaussian illusion.
On reading the book for the second time I kept asking myself "is this new to me?" and, if it is, "what is its relevance to me?". On reviewing one's own life to date, as the author recommends, it is clear that much of it has indeed been determined by high-impact unforeseeable events. This does not come as a great surprise - but then Taleb says that it never does, in retrospect! This contrasts with the chilling realization that there are almost certainly more such occurrences ahead. It is interesting to read explanations for why humans "don't know they don't know" they live in an extreme world but many, without realizing it, will already be familiar with psychological phenomena such as "platonism", "tunnelling", "confirmation bias" and the "narrative fallacy".
Interestingly, Taleb seems to miss what could well be the main reason why individuals "don't know they don't know": they just don't want to know they don't know they don't know! It seems to be a natural human reaction to put one's head in the sand when faced with the possibility of unforeseeable, high impact, possibly negative, events - particularly when they believe they can do nothing about them.
What, perhaps, is newer and more relevant to many is the fact that the professionals apparently rely on defective tools for analysing their particular piece of reality. Having some knowledge of the financial world and its questionable mathematical models, I can readily believe that many professionals - and even Nobel Prize winners - are led astray by the humble Bell Curve, as Taleb suggests. In fact the reasonably experienced small investor already has little faith in market "experts". On the other hand this same investor does not automatically transfer his scepticism to experts in other important fields, such as the social sciences, economics, environmental studies and military planning, where predictive errors can be far deadlier.
Besides peppering his text with the names and contributions of important thinkers - apparently a deliberate technique to achieve greater credibility - Taleb gives us some fascinating theory in the shape of non-linear relationships, the limitations of the Gaussian distribution, and the ability of so-called "power laws" to turn some Black Swans into Grey Swans. However this review stops a long way from demonstrating that life is largely determined by full-blooded (i.e. totally unpredictable) Black Swans.
Although I don't think that Taleb will make us see our lives in a totally new light it is important that he reminds us - in case 24-hour world news ever allows us to forget - that day-to-day affairs can be subject to unforeseen, and potentially devastating, modification. He also offers us the flip side: some ideas on how we can take advantage of positive Black Swans. It may be due to a lack of imagination but, not being a venture capitalist or a "quant", I couldn't immediately see measures of easy application in this area. Defence against Black Swans seems easier, namely diversification across very disparate fields. Taleb himself suggests a portfolio composed of up to 90% of extremely safe financial instruments (like Treasury bills) and as little as 10% in leveraged speculative bets like options (ideally involving "venture-capital style portfolios"). His general advise is more homely: learn to recognise undertakings exposed to positive and negative Black Swans, don't be narrow minded, seize opportunities, be wary of government plans, go to parties, etc.
Many interested in the impact of randomness will find the book a good, if fairly demanding, read. In the end, however, it is not entirely satisfactory. This has something to do with the fact that rather than be carried long by a limpid river of reasoning we are subjected to an avalanche of opinionative observations, some relevant, some less so. Some readers might also find many of the chapter and section headings irritating: "The Vagueness of Catherine's Lover Count", "How many Wittgenstein's can dance on the head of a pin?", etc. I suppose its all part of the relatively successful effort to make randomness fun. But although we'd enjoy seeing pompous academics and self-satisfied hedge-fund partners squirming with mice down their necks is it really necessary to rub the long-suffering French up the wrong way?
on 10 February 2013
I'm surprised that there are so many negative reviews for this book, I was expecting more positive reviews but they appear to be fairly evenly distributed over the ratings. This main idea in this book is the Black Swan which (as other reviewers point out) are unpredictable but high impact events. There are other related topics which are used mainly as support but are relevant as they illustrate why prediction in general is difficult and why it is difficult to prepare for such events. There are several nods to psychology (mainly Kahneman) as well as to philosophy (often to Russell but others too). There is an enjoyable narrative here which is peppered with many contentious comments that will enrage some and make others scratch their head. It isn't always obvious how much we should take literally and how much we need to simply take the meaning and then think on it ourselves. It is hard to distil difficult topics into a concise text: there will always be something overlooked.
What we have is an interesting narrative (ooh the irony) that depicts the failure of poor/ naive risk management over the last three decades of his professional career. He lambastes the use of Gaussian statistics and points out where such statistics are appropriate (games) and where they are not (real life). If you consider his position of living, nay... surviving, in an industry for 30 years where your view is that of an outsider then I think you should be able to understand the aggressive tone. Given this context you will see why the tone is not aggressive for the sake of being aggressive, unfortunately people who don't understand how far apart Taleb's thinking is from the conventional ways then the style and (perhaps) content will be unfairly dismissed as puerile. In an interview the author states that he has fought against the prevailing value-at-risk models. The structure of the book is fairly light as it wanders from topic to topic, it would be unfair to say incoherent as that isn't the case but rather the book is not highly structured like a textbook but rather like a novel.
There is little offered in the way of protecting oneself against a Black Swan, nor is there a strong explanation of how society can prevent/ resist Black Swans. However, I don't think that is quite the point of the book, the first step is learn how to understand what a Black Swan is then learn how to identify potential sources (hence become Grey Swans). Much of the book is epistemological (concerned with knowledge) rather than practical, it isn't a how-to guide for living.
I also think people that are unfamiliar with the financial lingo or economics terms may struggle to understand the author's use of particular anecdotes and references. Also his political or philosophical preferences won't be to everyone's taste which is another source of disagreement, Taleb states his preference for a society that allows failure, and one where a free market exists. Naturally, there is a lot of criticism of poor prediction at the within government and corporations, the author believes that such errors are compounded by over centralization in the related power structures where a small error can produce a large crack (a fragile entity: this doubtlessly upsets some readers who are keen central planners).
Readers who are not from a maths/ physics background may struggle with the copious use of mathematical terms. The book really isn't dense nor abstruse; I found it highly readable and at no point was I stuck but I'm coming from a maths/physics background, and I can see this being a problem for others. The concept of standard deviation is referred to but only loosely explained. A keen reader while search the internet to figure it out while the lazy or disgruntled (perhaps ideologically opposed to the book's content) will come to Amazon and write a one star review. The author could have made a bigger effort to explain these terms such that the book can have a wider readership.
on 28 October 2013
I was introduced to this book by a mathematician who advises on risk and who said he had revamped his ideas on what risk means. The black swan itself is a great metaphor for the kind of disaster that sends us all sideways or puts us out of existence. In the west we all thought all swans were white until explorers found black ones in Australia. Nothing in our experience could have led us to guess that black ones existed too. And so we take the present for granted, like turkeys enjoying the good life and not expecting any kind of dramatic change as Thanksgiving Day arrives. Black Swans in human life - like the 2007 financial crisis or September 11 - are rare, have significant implications and, once seen, appear as inevitable, or almost inevitable, consequences of what went before. Taleb goes on to explain how we humans like to convince ourselves that we can predict risk and manage it. How else would financial advisers be able to sell so many investment plans that then disappoint? Without such faith we would no longer put any faith in those long-term government predictions which we read all the time? Despite all his scepticism, Taleb outlines ways to spot some kinds of risks and how to reduce your exposure. He is the man behind Donald Rumsfeld's 'unknown unknowns' (aka Black Swans). However, Taleb labours his points amid some very tedious stories - and it takes some patience to wade through them.
The Black Swan: The Impact of the Highly Improbable has been a recent bestseller recently. Given the current economic turmoil, the lessons that Taleb imparts in this book are even more relevant. However, much of what is interesting and relevant is undermined by the author's attitude and dare I say, arrogance.
Let's start with the definition of a black swan. It is a rare (or outlying) event, which has a maximum impact and that with hindsight could have been predictable. What Taleb reiterates throughout the book is that we should not confine ourselves to the realm of the usual and predictable. Fair enough.
Taleb then goes on to disparage Gaussian statistics and the reliance that is placed on such statistics, particularly in financial or economic matters. Given that Taleb is an ex-trader himself, I found it highly interesting that he doesn't detail any of the modelling methods that he employed. He refers to himself as a skeptical empirist, someone who is not easily convinced.
However, as a physicist (though not working as such anymore) I found his bashing of the bell-curve to be tedious. Most good scientists are well familiar with the limitations of such statistics and do not place any real credence in them. However, judging from the author, you would gain the impression that all of the world is founded upon these methods, and it's just waiting to come crashing down.
In short, I did enjoy his concept. Black swan events should happen and we should not ignore them when they do. Life is a blend of random and predictable and to ignore the former in favour of the latter is hardly a clever thing to do.
I did not enjoy the book because of the Taleb's attitude. His sense of superiority is astounding, especially when you consider that he rails so often against the arrogance of those who think they know it all.
on 23 January 2009
As many have said before me, Taleb's arguments are valid and important (to financial markets and to life more generally). As has also been said, they are not original, the style is hectoring, the conceit is fabulous, and the book is therefore tiresome.
However, if you are tempted to read the book, might I strongly suggest you read his previous book, Fooled by Randomness, instead? The same message is put forward more succinctly, with more humour, with less conceit, and with more interesting little side stories. It even explains the Black Swan, so you don't miss out on that.
The message is important, but this book isn't.