Somehow, I forgot a few books you should definitely check out:
The Black Swan and Fooled by Randomness by Nassim Taleb
The Misbehavior of Markets by Bernard Mandelbrot
All three deal with the same problem: faulty probability models being used by financial wizards, who assumed that they could quantify risk along normal distributions. The Mandelbrot is the most technical (he's the mathematician who brought you fractals), so read the Taleb first.
Here's an article about Nassim Taleb's investment strategy by Malcolm Gladwell, which is a little light but gives you a rough idea of what he's getting at.
Copyright © 2010 by The Atlantic Monthly Group. All rights reserved.
Every time I mention Taleb's "Fooled by Randomness" to actual options traders, they're pretty derisive. I wasn't a fan of the book myself. It's heavy on speculation and light on hard data. The traders I talk to say that risk is usually accurately priced into the models. It's worth nothing that a hedge fund Taleb founded a few years back went broke pretty quickly.
I believe M. Mandelbrot's first name is Benoit
faulty probability models being used by financial wizards...
----------------
Well, AIG was no "long-term capital management", the firm with the most Orwellian name of the last century, possibly.
All's to say, we need more than just to point fingers at "wizards".
Has someone written a book about CEOs and boards who take a hands-off, no-questions-asked-or-limits-imposed-approach, so long as the dollars are flowing in, hand-over-fist?
Pathological monsters! cried the terrified mathematician
Every one of them is a splinter in my eye
I hate the Peano Space and the Koch Curve
I fear the Cantor Ternary Set And the Sierpinski Gasket makes me want to cry
And a million miles away a butterfly flapped its wings
On a cold November day a man named Benoit Mandelbrot was born
His disdain for pure mathematics and his unique geometrical insights
Left him well equipped to face those demons down
He saw that infinite complexity could be described by simple rules
He used his giant brain to turn the game around
And he looked below the storm and saw a vision in his head
A bulbous pointy form
He picked his pencil up and he wrote his secret down
Take a point called Z in the complex plane
Let Z1 be Z squared plus C
And Z2 is Z1 squared plus C
And Z3 is Z2 squared plus C and so on
If the series of Z's should always stay
Close to Z and never trend away
That point is in the Mandelbrot Set
Mandelbrot Set you're a Rorschach Test on fire
You're a day-glo pterodactyl
You're a heart-shaped box of springs and wire
You're one badass fucking fractal
And you're just in time to save the day
Sweeping all our fears away
You can change the world in a tiny way
Mandelbrot's in heaven, at least he will be when he's dead
Right now he's still alive and teaching math at Yale
He gave us order out of chaos, he gave us hope where there was none
And his geometry succeeds where others fail
If you ever lose your way, a butterfly will flap its wings
From a million miles away, a little miracle will come to take you home
Just take a point called Z in the complex plane
Let Z1 be Z squared plus C
And Z2 is Z1 squared plus C
And Z3 is Z2 squared plus C and so on
If the series of Z's should always stay
Close to Z and never trend away
That point is in the Mandelbrot Set
Mandelbrot Set you're a Rorschach Test on fire
You're a day-glo pterodactyl
You're a heart-shaped box of springs and wire
You're one badass fucking fractal
And you're just in time to save the day
Sweeping all our fears away
You can change the world in a tiny way
And you're just in time to save the day
Sweeping all our fears away
You can change the world in a tiny way
Go on change the world in a tiny way
Come on change the world in a tiny way
The take away I got from Taleb's book was that while you and split the cost of lunch between two people, you can't split shooting yourself in the head between to people.
What Would Graham Do?
A positive investment analysis in the WSJ by Arnold Zweig:
http://online.wsj.com/article/SB122368241652024977.html
"Strikingly, today's conditions bear quite a close resemblance to what Graham described in the abyss of the Great Depression. Regardless of how much further it might (or might not) drop, the stock market now abounds with so many bargains it's hard to avoid stepping on them. Out of 9,194 stocks tracked by Standard & Poor's Compustat research service, 3,518 are now trading at less than eight times their earnings over the past year -- or at levels less than half the long-term average valuation of the stock market as a whole. Nearly one in 10, or 876 stocks, trade below the value of their per-share holdings of cash -- an even greater proportion than Graham found in 1932. Charles Schwab Corp., to name one example, holds $27.8 billion in cash and has a total stock-market value of $21 billion.
Those numbers testify to the wholesale destruction of the stock market's faith in the future. And, as Graham wrote in 1932, "In all probability [the stock market] is wrong, as it always has been wrong in its major judgments of the future."
In fact, the market is probably wrong again in its obsession over whether this decline will turn into a cataclysmic collapse. Eugene White, an economics professor at Rutgers University who is an expert on the crash of 1929 and its aftermath, thinks that the only real similarity between today's climate and the Great Depression is that, once again, "the market is moving on fear, not facts." As bumbling as its response so far may seem, the government's actions in 2008 are "way different" from the hands-off mentality of the Hoover administration and the rigid detachment of the Federal Reserve in 1929 through 1932. "Policymakers are making much wiser decisions," says Prof. White, "and we are moving in the right direction."
Maybe people should read Benjamin Graham.
"The Shock Doctrine" by Naomi Klein.
Try reading Henry George's Progress and Poverty. Back in 1879, George explained how booms and busts in land prices are responsible for triggering panics and industrial depressions, and why land is particularly subject to speculative frenzies. It certainly seems to fit what's been happening 120+ years later.
Successful option trading often involves determining when the premiums are wrong. In any case exchange traded stock options are in no way directly responsible for this mess. They are responsible indirectly to the extent they are misused, often by making players imagine they are hedged properly but end up being over leveraged.
Assets of all stripes have been inflated by excess credit, leverage. If credit was infinite then one might logically conclude that more leverage would solve this problem. Good luck with that.
Benoit, not Bernard.
The Black Swan is great. The problem was that many many people, not just schmucks like me but people like Volker and Henry Kaufman saw flocks of black swans more than a decade ago. The people who were paid not to see them didn't. Then there were people like Greenspan who were just ignorant enough of history to believe they were masters of it. Finally there were the Paulson's who knew it would happen and have positioned themselves accordingly to garner even more power going forward. Sure whey will lose some money but power is better.
In other words the credit market collapse wasn't a black swan. A banking collapse isn't a black swan. It's what banks always do. It's a pure white swan.
ChrisV,
Just as a matter of curiosity, how many of those derisive options traders are still solvent?
I am a little dissapointed that you dont have "When Genius Failed" in your list
Taleb's books are definitely worth a read and I think about Fooled by Randomness all the time. That being said, Taleb is a pretty poor writer and almost never sticks with a given thought to completion - he even wanders aimlessly during the course of single paragraphs. The editing of Fooled by Randomness was abysmal - too bad because a better editor could have turned that book into something truly indispensable.
On the other hand, The Shock Doctrine by Naomi Klein was suggested by another reader - now Klein can write. She is clear, concise and very intelligent. I concur with that suggestion - it's an important read for the times.
"When Genius Failed" was a good read, and I've wanted to get the The Black Swan for some time now.
In our current disaster, the black swan in question is credit default swaps, and their unregulated nature.
(On another note, what happened to The Atlantic's format. Colors look a wee bit vivid).
Best quote in that article:
But if anything completely out of the ordinary happens to the stock market, if some random event sends a jolt through all of Wall Street and pushes G.M. to, say, twenty dollars, Nassim Taleb will not end up in a dowdy apartment in Athens. He will be rich.
Wonder how many GM puts Taleb's company was holding onto.
Sancho -- the issue with Taleb and option models is not one of the solvency or lack thereof of option traders.
It is true and easily verified that the implied volatilities of out of the money options imply strong heavy tails and non-normal/non-Gaussian investing patterns. There is a kind of quirky way in which multiple Gaussians are used rather than a single Gaussian that goes by the name "volatility smile". Any old fella off the street can tell you that such "mixture models" and/or "extrapolations in volatility coordinates" can capture the kinds of issues Taleb wants captured. There are quibbles of model complexity/parsimony, coordinates, mathematical taste, and other subtleties, but actual option traders worry about those things, too, explicitly or implicitly.
So, the issue is quantitative not qualitative. Taleb might say the smile or smirk is "not smirky enough", but he could easily be wrong about that. The reason his fund went bust so quickly is that, counting transaction costs and also counting the limited terms options were available for, um, yeah, maybe they _were_ smirky enough for their duration. Maybe given all the money they were losing and would have lost up to the present (had they stayed in business), they would _still_ be negative even given the past month's wild market motions. Or at least no more net positive than the typical options trading fund. No one knows, but at least under the money-where-his-mouth is test, Taleb failed and should not be mouthing off his I-told-you-so's.
At least under the patience constraints of his investors, Taleb should have known he'd have trouble -- that he might need to wait 5 or 10 years for his black swans, not 2. I mean, betting his fund on *near term* "grey" swans is just as stupid as all the people he criticizes for ignoring long-term black swans to attempt to adapt the lame metaphorical terms.
As for options traders winning or losing around now, you should probably expect that they are doing well. The VIX is at an all time high and volume and volatility usually drive their profits, not directional moves of the markets. Traders that had a net short volatility bias before all this started and were unable to flip might have gotten caught off guard by recent volatility of volatility. Really smart houses could have used options on the VIX to control their exposure to such things, though....In any event, my understanding is that most options traders are doing quite well these days. Most traders are not volatility/options traders. So you would not know this from the news { nor is most money made in the business done by options trading so, no, it can not compensate for other losses in large i-banks with mortgage holdings }. It is easy for news coverage of this topic to be misleading.
So, maybe millions of short term plays are a better way to go than Taleb's hundreds of wild move plays...Empirically, there sure seem to be a lot more of those funds staying in business.
Personally, I think Taleb's points are mostly either obvious or vacuous and seldom the most useful way to conceive a situation and often poorly expressed, but as with Gladwell, he has a way of giving weak metaphors to the unmathematical masses that they then carry with them in bad applications. My mind may be muddled with a perfect storm of black swans tipping points in favor of Obama some other such metaphorical medley misapplied.
A book review of Misbehavior of Markets with which I largely agree: http://www.bayesianinvestor.com/blog/
woops, I meant to use the permalink: http://www.bayesianinvestor.com/blog/index.php/2008/10/05/misbehavior-of-markets/
All three deal with the same problem: faulty probability models being used by financial wizards, who assumed that they could quantify risk along normal distributions. --MM
you're famous 'smart guys' aka financial wizards, forgot the first, Elementary, lesson in Econometrics: Nothing equals Anything. Usually said after the caveat: you'll learn to be proficient in Mathematics, but remember: Nothing equals Anything.
after seeing Mandelbrot's 'fractals'(produced in a Static environment), anyone who then thought that they could predict the endless permutations/gradations of multi-variate(and still incomplete v. *Reality) equations, in a Fluid setting, was either: Stone-Cold Stoopid to a Statuatory degree, or a Post-Modern Snake Oil Salesman. Take the Quantum Bet: Both, at once.
MM is right, though, in pointing out those Books, they're excellent. But, just like the MSM, MM thought all was right with the World, until it broke, then: "Breaking News...Film @ 11."
Truly, a Day late, and a Dollar short.
MM, you need new handlers, your pro-Gaming is too painfully obvious.
IME the number of people in the investment business who pay serious-sounding lip service to 'The Black Swan' is enormous. The number of people who actually -- before this summer, at least, and probably still even now -- grasp its implications for what they do every day is tiny. It's definitely one of those books that is often bought (and displayed) but seldom read. Taleb is not a particularly skilled writer but it's not that hard a slog.
I'm just wondering how long it is going to take for Megan to apologize to Paul Krugman far all the nasty things she said about him and his predictions about the current financial crisis?
Either written with irony, or without self-awareness.
I would guess the latter.
For those who didn't catch the commenter's reference above, here is a link to Jonathan Coulton's song about Benoit Mandelbrot, "Mandelbrot Set". While you're at Coulton's site, check out his killer folk cover of Sir Mix-a-Lot's "Baby Got Back".
Don the Libertarian Democrat,
This company is also trading for less than its cash on hand, U.S. Energy Corp (USEG). Doesn't have current earnings, but probably will within 6-9 months, and it has plenty of cash to see it through until then. I've written about it in detail on my site, if you're interested, or you can do some digging on it on your own. There are also some bargains among larger stocks too. I'd just stick to the cash-rich ones that aren't dependent on the debt markets for financing or on the over-extended U.S. consumer for their sales.
btw, this cat:
Sancho -- the issue with Taleb and option models is not one of the solvency or lack thereof of option traders.
It is true and easily verified that the implied volatilities of out of the money options imply strong heavy tails and non-normal/non-Gaussian investing patterns. There is a kind of quirky way in which multiple Gaussians are used rather than a single Gaussian that goes by the name "volatility smile". Any old fella off the street can tell you that such "mixture models" and/or "extrapolations in volatility coordinates" can capture the kinds of issues Taleb wants captured. There are quibbles of model complexity/parsimony, coordinates, mathematical taste, and other subtleties, but actual option traders worry about those things, too, explicitly or implicitly.
So, the issue is quantitative not qualitative. Taleb might say the smile or smirk is "not smirky enough", but he could easily be wrong about that. The reason his fund went bust so quickly is that, counting transaction costs and also counting the limited terms options were available for, um, yeah, maybe they _were_ smirky enough for their duration. Maybe given all the money they were losing and would have lost up to the present (had they stayed in business), they would _still_ be negative even given the past month's wild market motions. Or at least no more net positive than the typical options trading fund. No one knows, but at least under the money-where-his-mouth is test, Taleb failed and should not be mouthing off his I-told-you-so's.
At least under the patience constraints of his investors, Taleb should have known he'd have trouble -- that he might need to wait 5 or 10 years for his black swans, not 2. I mean, betting his fund on *near term* "grey" swans is just as stupid as all the people he criticizes for ignoring long-term black swans to attempt to adapt the lame metaphorical terms.
As for options traders winning or losing around now, you should probably expect that they are doing well. The VIX is at an all time high and volume and volatility usually drive their profits, not directional moves of the markets. Traders that had a net short volatility bias before all this started and were unable to flip might have gotten caught off guard by recent volatility of volatility. Really smart houses could have used options on the VIX to control their exposure to such things, though....In any event, my understanding is that most options traders are doing quite well these days. Most traders are not volatility/options traders. So you would not know this from the news { nor is most money made in the business done by options trading so, no, it can not compensate for other losses in large i-banks with mortgage holdings }. It is easy for news coverage of this topic to be misleading.
So, maybe millions of short term plays are a better way to go than Taleb's hundreds of wild move plays...Empirically, there sure seem to be a lot more of those funds staying in business.
Personally, I think Taleb's points are mostly either obvious or vacuous and seldom the most useful way to conceive a situation and often poorly expressed, but as with Gladwell, he has a way of giving weak metaphors to the unmathematical masses that they then carry with them in bad applications. My mind may be muddled with a perfect storm of black swans tipping points in favor of Obama some other such metaphorical medley misapplied.
Posted by c | October 12, 2008 12:09 AM
obviously, knows what he's talking about. really, surprised to see him on these threads..
MM, if you want someone knowledgable to refer to/quote, you may want to keep him in mind..
dantonj, the best comment I remember reading about Paul Krugman is that he's successfully predicted nine of the last two recessions. I guess he's up to being ten for three now.
I'm just wondering how long it is going to take for Megan to apologize to Paul Krugman far all the nasty things she said about him and his predictions about the current financial crisis?
Feel free to recapitulate any and all of Krugman's predictions about the current financial crisis (i.e., those that can be legitimately seperated from Stopped Clock Effect), so we can review actual evidence.
Also, IIRC very few people around these parts are in the habit of saying nasty about Krugman in his capacity as an economist. As a political commentator, or as a political commentator posing as an economist, then yes. But not as an actual economist.
In addition to the comment that the Mandelbrot of fractal fame is more usually named "Benoit" than Bernard, I think the more important point is that in no way did Benoit Mandelbrot invent fractals. The Mandelbrot set has proved a particularly fruitful source of images when plotted and coloured in various way, and provides certain insights into the nature of another set of fractals, known as the Julia sets, but that is about it.
Krugman has never, ever, not once, done a Bloggingheads. He never posts vegan recipes. He never cites to everything Tyler Cowan ever wrote. He is not tall. In short, he is the least qualified to win the Nobel Prize in Econoblogging. Shame on you, Sweden.
I agree with HungChad. Taleb is a weirdly self-absorbed and unfocused writer. His point, that models are incorrect, is trivial and has been well-known for decades. A lot of financial research over the past few decades has been to explain and deal with the issues of kurtosis in financial probability models (i.e., "fat tails"). Taleb merely acknowledges their existence, without offering much in the way of an explanation, or much in the way of useful prescriptions.
A better book along similar lines is "Plight of the Fortune Tellers" by Riccardo Rebonato.
The Gladwell article was really fun. The idea of setting a large probability of small, manageable losses against a small probability of enormous wins is quite reminiscent of the equally entertaining (but absolutely *not* to be used for investment purposes...) Thirteen Against The Bank.
Once again Europe is trying to influence the American election. Giving the Nobel to McCain basher Krugman is such an obvious ploy to sway votes towards Obama that it is laughable.
Don (Megan's fellow singular libertarian Democrat) and DaveinHackensack seem to be the only peole living in October 13. Today and for the rest of this week we are likely to be offered more and better bargains on the world's stock markets than at any time in the next decade or three. Megan is real sick and needs bad to recuperate; so she has timew for reading this week. But the rest of you? Don't you really want to be rich?
Ugh, they're all so tediously written.
At least under the patience constraints of his investors, Taleb should have known he'd have trouble -- that he might need to wait 5 or 10 years for his black swans, not 2.
c, if you are trying to lower my opinion of Wall Street, its inhabitants, and its investors, you are doing a fine job.
You must be aware that the traditional left-wing criticism of the financial markets is that they are controlled by short sighted cretins.
And you just agreed with those critics.
(Amusing to note that the vaunted New Yorker subs desk didn't catch "igon value" for "eigenvalue" in the Gladwell piece, though.)
Giving the Nobel to McCain basher Krugman is such an obvious ploy to sway votes towards Obama
Even if this weren't idiotic on the merits -- see: Tyler Cowen's thread, its comments chock-full of people who clearly have no respect for Tyler Cowen's opinion here -- it's idiotic on the politics, given Krugman's coolness towards Obama's economic proposals during the primary.
Not about misapplying normal distribution, there are plenty of ways to modify the shape of a normal distribution if you have a sound theory, know what will break your model, and keep an eye out.
The recent problem was that they didn't expect default risk to change. After 2005 incomes declined for a lot of people and non-mortgage expenses went up. This meant budgets were persistently tight; the financing people had availible to endure shocks (home or car repairs etc.) couldn't help with a couple years of higher than expected living expenses. This totally skewed people's utility functions and meant that models used to determine which mix of assets to own in previous years no longer matched the new reality.
Mandelbrot isn't exactly known as a technical wizard of a mathematician. His ideas are influential more because of creativity and breadth of applications than sharpness. A lot of the hard work for which he will eagerly take credit was done by his graduate students. He's well-versed enough to invoke lots of technical ideas, but it's unclear to me how good his command of them is, so if the book seems confusing or too "technical", it may not be because of reader error. Just a grain of salt to take with you.
"Personally, I think Taleb's points are mostly either obvious or vacuous and seldom the most useful way to conceive a situation and often poorly expressed, but as with Gladwell, he has a way of giving weak metaphors to the unmathematical masses that they then carry with them in bad applications. My mind may be muddled with a perfect storm of black swans tipping points in favor of Obama some other such metaphorical medley misapplied."
This sums up my thoughts about Taleb perfectly. I have read three of his books and found his conclusions overstated and often wrong.
I am also very skeptical of his investment strategy of being long options as I have seen several other actual academic articles show that a short options strategy is the one that is a long term winner.
His best book was his first on the technical details of running an options book "Dynamic Hedgeing". "Black Swan" and "Fooled by Randomness" were too much like thier cousins "Tipping Point", "Blink", and "The World is Flat" in that they took a simple idea that is in some ways true some times and rammed it into every concievable situation they could think of for 200 pages.
Fooled by Randomness is an unreadable book, given to using too-cute parables and lacking a unifying theme. It's too bad, because the book is based on a variety of important and intriguing ideas, such as biases in decision-making, the limits of Gaussian distribution and important tenets of scientific epistemology, like ideas from Popper, Hume and so forth. Better explanations of similar ideas can be found in the recently published "Drunkard's Walk" and Ariely's "Predictably Irrational." Taleb also has a recent, non-technical piece on the Edge webste that describes the limits of basing models on assumptions of Gaussian distribution [http://www.edge.org/3rd_culture/taleb08/taleb08_index.html]
This was an interesting time to read Brian Doherty's "Radicals For Capitalism". I never really knew much about the Austrians before, but reading about them as this mess came to pass was rather illuminating.
You okay, Megan hon'? You've been awful quiet.
Interestingly, I was Mandelbrot's research assistant in college for two years, and Nassim Taleb's TA in grad school when he taught Mathematical Finance at NYU.
I think both guys provide useful critiques of how probability theory is used in finance, but I think they're off-base in some of their most extreme rhetoric, which is the stuff that gets them published and large speaking fees.
Look, sophisticated options traders and almost all quants know that the stock market is not log-normal. The models they use aren't just basic geometric Brownian Motion (the model assumed by Black-Merton-Scholes). Most are fairly sophisticated models that attempt to exhibit the fat tail behavior seen in the markets. To be sure, Mandelbrot's paper with Samuelson in the 60's (maybe 70's?) convinced people that indeed, asset prices did not behave log-normally. And Taleb has done important work (his Dynamic Hedging book is still an industry standard). But that doesn't mean we need to accept their claims even when as extreme as they are.
Everyone knows the models are imperfect. But making them perfect, using current knowledge, would mean that you could no longer price options in the real world. Also remember, it's not people's equity and FX option books that have blown up. CDO's, MBS, CMBS, RMBS, etc are not priced according to the models criticized by Taleb and Mandelbrot. Most of those are priced using static models rather than dynamic ones (since dynamic models are really really really friggin hard to make workable). In fact, banks' option books have done quite well in the recent turmoil.
Everyone knows the models are imperfect. But making them perfect, using current knowledge, would mean that you could no longer price options in the real world.
If everyone uses the same (slightly) wrong models, doesn't that make them likely to all hit trouble at the same time?
Or is that a feature, rather than a bug? If you hit trouble on your own, Uncle Sam is less likely to bail you out...
Best to make the same mistakes as everyone else.
As far as options go no one even "Uses" the models in the sense that people think.
The Black-Scholes-Merton option model mainly provides a language and a framework for thinking about options. No one takes any of the assumptions as originally proposed seriously. And each option trader uses his own judgment in adjusting those assumptions.
Option traders quote the price of options in implied volatilities. This is the volatility that you would put into the model to the value of the option given all the other parameters. This value trads just like the prices of stocks trade and every option has an implied volatitlity that varies from stock to stock and with the duration and strike of a given stock.
Whether the current volatility (price) of an option is a buy or a sell at a given level is a judgement call by each individual trader and they all have differing opinions.
Now MBS are different, but the opinions can and do vary about the assumptions underlying those models as well. Some people did diagree with the assumptions about default probablities and avoided most of the MBS mess or made money off it at least in the early stages (Goldman).
As far as actual trading, models are almost always a shorthand for valuing and hedging and good traders and quants know thier shortfalls.
Where you tend to get in trouble is when rigid risk management regulations are put in place based on those models. You can never substitute a hard and fast rule for the experience of a market participant. But the market participants don't have as much of an incetive to manage risk they own the traders option (big bonus if I win, only fired if I lose) so they have an incentive to take risk.
The goal of the risk manager is to keep the company out of trouble, but he can only really do so by writing down a rule book for traders and the rules are often based off of probablistic models. As long as the traders play by the rules they can trade freely. The problem is that the traders can figure out how to game the rules in millions of different ways by taking risk that they know about but that are not anticipated by the rulebook.
I think this has been the most impressive comment section ever. Unfortunately, my comment may be a regression to the mean. Just in watching the markets though and reading comments about 'Black Swan events' being more 'normal' than implied by a normal distribution reminds me of Einstein's comment that 'God doesn't play with dice.' Unfortunately, Isaiah didn't seem to mention this but G-d seems to play with multiple pairs of dice. In which case the question may be: what is the probability that in an unkwown "a" rolls of 2 dice the least probable 2.78% for rolls of 2 dice will occur in one pair, etc.?
What I've learnt from this is that the market can stay rational longer then Taleb can stay solvent.