The Formula That Killed Wall Street 561
We recently discussed the perspective that the harrowing of Wall Street was caused by over-reliance on computer models that produced a single number to characterize risk. Wired has a piece profiling David X. Li, the quant behind the formula that enabled the creation of such simple risk models. "For five years, Li's formula, known as a Gaussian copula function, looked like an unambiguously positive breakthrough, a piece of financial technology that allowed hugely complex risks to be modeled with more ease and accuracy than ever before. With his brilliant spark of mathematical legerdemain, Li made it possible for traders to sell vast quantities of new securities, expanding financial markets to unimaginable levels. His method was adopted by everybody from bond investors and Wall Street banks to ratings agencies and regulators. ... [T]he real danger was created not because any given trader adopted it but because every trader did. In financial markets, everybody doing the same thing is the classic recipe for a bubble and inevitable bust."
Citation, please (Score:5, Interesting)
Citation? Booms and busts are caused by, respectively, expansion and contraction of the money supply (usually in the form of bank credit), often accompanied by manipulated interest rates. The formulas used by lots of investing firms could cause clusters of errors, but the extent of types of companies (and governments) affected points to a more Austrian-style, systemic boom/bust rather than a single-(important-)sector miscalculation.
Picking up pennies in front of bulldozers (Score:5, Interesting)
That Gaussian curves are a poor model for unlikely events has been known for quite some time. This is best explained by Nassim Taleb in the following books:
His main thesis is that the markets are essentially random and are basically impossible to predict in any meaningful way. Further there are unlikely unknown unknowns can cannot be predicted until the they occur, usually with disastrous consequences.
Re:Nothing wrong with models. (Score:5, Interesting)
Re:Citation, please (Score:3, Interesting)
Classical economics cannot explain what is happening right now. It's without precedent. There is a little graph I would like to show [msn.com] you...
It's interesting to note the near exponential shape of the graph pre dot com bust era, and how the exponential part resumes around 2005. Now, imagine the impact on everyone with money to invest, from corporations to banks to retirement and pension funds faced with a choice. You can earn 4% or less, per annum, in bonds or (LAUGH) CD's, etc. OR you can put money on the stock market. That's one hell of an "opportunity cost" if you don't - because everyone else is making out like a bandit. The stock market is unstoppable.
In fact, the only OTHER "safe" place to put your money is real estate...
Both of them went bust at roughly the same time. Co incidence? No, they were intertwined from the beginning, because they were the "safest" "surest" bets, and that's where all the wealth was going. So according to supply and demand, if too much money was chasing these "goods", the price moved up accordingly. However these two retracements have wiped out the present AND FUTURE wealth of most of the nation, because everyone was BANKING on the fact that their stocks, 401(k) or home was going to see them through retirement. Welcome to reality - the money is gone (because the demand is gone), and we're not finished yet. The graph still points STRAIGHT down. Something HUGE has to happen to change that. Most people thought it would be a new president, but now we know that's not the case.
Similar to a hedge fund? (Score:3, Interesting)
This story reminds me of "Long-Term Capital Management" story back in the late 1990's.
http://en.wikipedia.org/wiki/Long-Term_Capital_Management [wikipedia.org]
These guys did the EXACT same thing using computer models to predict what funds they should be investing in so that they never have a loss ...
Unfortunately, they were bailed out, but folded in 2000.
http://www.geocities.com/eureka/concourse/8751/jurus/hf100203.htm [geocities.com]
There was a PBS special about these guys and the computer models they used.
http://www.pbs.org/wgbh/nova/transcripts/2704stockmarket.html [pbs.org]
brilliant spark of mathematical legerdemain (Score:3, Interesting)
My grandfather woulda thought this guy was a Red infiltrator. Good job if he was.
Re:the formula that killed wall street: (Score:4, Interesting)
Greed is a motivator. Greedy people will work hard to acquire money. Capitalism & free enterprise allow a society to harness greedy people for positive ends like the creation of jobs to produce valuable goods and services. This is a good thing. Unfortunately, greedy people are not necessarily *smart*. And even the smart greedy people are not necessarily *correct* when they do things a particular way.
The story sums it up nicely - this formula oversimplifies a complex market creating a classic bursting bubble. There's an economist named Taleb http://www.fooledbyrandomness.com/ [fooledbyrandomness.com] lecturing about how the market will basically always be more complex than you think.
The best part about his message is in not trusting your data too much. I think of this every time people start talking confidently about geoengineering. We don't know as much as we think we do.
Re:Nothing wrong with models. (Score:5, Interesting)
it does not matter what model you use. Apparently they all created virtual worlds in big numbers (total value of derivatives and such is few times more than summed up gross domestic product of all countries on our planet) - this had to crash independently of the model - problem being that they used the same one. in other words: if all sheeple use the same model of reality then to make profit you need to use different one. Or to say it yet differently: if all sheeple do the same they create the bubble. nature of bubbles is that they burst when they reach physical limits of the stuff of which they are made. In our case it was human gullibility.
Re:Nothing wrong with models. (Score:5, Interesting)
>>There is nothing wrong with using a model. Models are good.
Not in economics, they're not. The book Black Swan, which should be read by anyone interested in this topic, says that the hideous lie is that people claim that "they're better than nothing", when, in fact, they're worse than not having any model at all.
The LTC crash was caused by the founders (Nobel Laureates in Economics) having a model to quantify risk. IIRC, they used some sort of guassian model, taking the standard deviation of price movement as "risk". (http://en.wikipedia.org/wiki/Black-Scholes#Black.E2.80.93Scholes_model) This of course looked good until, quite suddenly, it wasn't and there was an event that their model predicted shouldn't have happened within the lifetime of the universe (that's the problem with using gaussians instead of cauchy curves or other fat-tailed distributions) and the company crashed and burned, and did a lot of collateral damage as well.
From the wikipedia article on LTC (http://en.wikipedia.org/wiki/Long-Term_Capital_Management): Merrill Lynch observed in its annual reports that mathematical risk models, "may provide a greater sense of security than warranted; therefore, reliance on these models should be limited."
Re:the formula that killed wall street: (Score:5, Interesting)
yes, i completely agree with you. the focus on quarterly earnings is representative of "short-termism" everywhere, which is usually detrimental to long term value preservation.
i guess what i should have said is that greed is not going anywhere. harness it when you can and don't be surprised when it causes people to do things that harm others.
Re:Picking up pennies in front of bulldozers (Score:4, Interesting)
Otherwise, it's all just blind gambling (which it isn't far off, anyway).
The advent of full on algo trading means that random events in the market have the ability to wipe out tons of capital because the models predict (e.g.) a global crash when it's just a blip. (Extreme example)
The other part of the problem is that traders are nowadays just glorified clerks in that all (well, 90%+) of the actual calculation and predictive work is done by complex platforms (or Excel), so they don't really care or have exposure to the real risks behind their trading.
Coupled with the huge bonuses they used to get (I'm in London where bonuses are being denied; is it the same elsewhere?) as long as they showed *quantity* of trades, it was always a recipe for disaster.
Diversity is good, 'mmkaay? (Score:4, Interesting)
Any sufficiently complex system should be heterogeneous, so that not all parts of the system can fail due to the same flaw.
Any homogeneous system will inevitably be at greater risk of failure due to a flaw in the common "gene pool" so to speak.
Biology, computers, economics, politics... I could go on.
Re:Economic Stimulus (Score:3, Interesting)
In China, they're using this slack time to upgrade the infrastructure, closing down old inefficient factories and building new ones with government CASH. Who's winning this round?
Re:Picking up pennies in front of bulldozers (Score:2, Interesting)
Re:Economic Stimulus (Score:5, Interesting)
In China, they're using this slack time to upgrade the infrastructure, closing down old inefficient factories and building new ones with government CASH. Who's winning this round?
Not the millions of migrant chinese workers who have lost their jobs, which will probably also cause civil unrest. Also, the Chinese holding trillions of dollars in U.S. treasuries will also be slightly annoyed when the U.S. government inflates away their debts.
Finally, the vast majority of China's stimulus package was already announced before this major recession. You have the order backwards.
Two sides to every trade (Score:3, Interesting)
So if the Street were all one way (hypothetically) then the
counterparts are the otherway.
The genesis of this debacle lies as much with the buy side
- pension funds, mutual funds, etc who were willing to buy
anything so long as they got a pickup of 15-25bp over the
comparable treasuries. In effect, they asked for this
stuff and they got it.
As to VaR - its a great way to model relatively stable
markets and to quantify short term risks of a large move
based on recent historical returns, volatility and asset
correlations. It's not meant to predict trends nor to
quantify 'what if the market for X tanks every day for Y
months'. Thats what managers and traders are for - to
realize there has been some change, perhaps fundamental,
which will have a long term negative effect on their
positions and to take what action is necessary to reduce
that risk. Instead, they froze.
Oh Please... (Score:5, Interesting)
There's nothing advanced or innovative about a gaussian copula. It's a very simple mathematical trick, it doesn't say anything about finance in itself. It's a programming trick to go from a uniform distribution on a cube (easy to generate, run rnd() for each coordinate) to a multivariate gaussian with a specific covariance matrix. The way to do it is cholesky decomposition. This is OLD stuff.
Li's paper is a clever way to measure default correlation using correlation matrixes from asset returns. It's quite clever, and yes it's a pretty good model (more on that later)
This is not journalism, this is a bit of shit where the author decided having an "evil formula" would be cool. Look there's an "equal" sign, how can they be so sure... pffffffffffffffff.
I said it was a good model, yet it's been proven wrong hasn't it? Well, first of all, what has been shown to be wrong is the guesstimate of correlation that was input into the model. G.I.G.O
Plus, if you price a fixed income product and it produces higher than market return, you will borrow short term funds to invest them in it. In a free market that quickly drains the pool of saving and raises short term interest rate. Sure you end up losing money but no catastrophe. In a federal reserve system, well the short term rate stays what the fed says it should be and everyone piles on the arbitrage, creating sky high leveraged position.
Yeah the formula can be misleading, but for a true catastrophe, you need a federal reserve.
Let's not forget (Score:3, Interesting)
Re:Nothing wrong with models. (Score:5, Interesting)
>>>they all created virtual worlds in big numbers - [the real world] had to crash independently of the model
Maybe we should invent a game for these bankers. World of Real Estate - where the goal is to get as many poor people into as many houses as possible, without investors learning the real housing value is only half the retail value. That way they can watch their virtual bubble go "boom" without affecting the rest of us in the real world.
Re:the formula that killed wall street: (Score:3, Interesting)
I would just hope that we can temper our greed
Huh? In other words, if I have the intellect and the good fortune to make an extra dollar, I shouldn't? We're not talking necessarily about evil pyramid schemes here. We're talking about the aggregate effect of every day investors that see the potential of making literally an extra penny per share by taking a course of action.. Since they're essentially gambling, they've taken incredible losses over the months (any given set of months), and thus are more and more desperate for every little bit that can count. Couple this with the fact that many investors are rated based on their ROI performance - they are not acting out of greed, they're acting out of self preservation for their job.
So are you saying, the aggregate masses should be a little less greedy? Then that 1% net profitability for which gambling greed drives them should simply not play.
Hey, I'm all for it man. I think Mutual funds and Hedge funds were the beginning of the end of Capitalism as we know it. Lets go back to Direct investment where people have a stake in their stock. Make retirement funds for the middle class purely interest bearing again. I'm not even being sarcastic - all my retirement is in inflation based securities. I'm in disbelief of the 401k mania over the past 15 years - and scold republicans for wanting to replace S.S. with 401ks.
Models had nothing to do with it (Score:3, Interesting)
The contribution of mathematical models to the present crisis has been vastly overblown. The breast-beating and mea culpas from the likes of Derman and Wilmott are self-flattering: after all, if you caused the problem, you must be important! In reality, the quant is at the bottom of the pecking order on most trading floors. The people who trafficked in securitized garbage did so not because they were fooled by their models, but because they were paid to. You can't tell me that the guy who lent $750,000 to a strawberry picker with $14,000 income would have thought it a good idea if he was lending his own money.
Contrast this to LCTM, which really is an example of quants gone wrong: those guys had so much faith in their models that they not only put their own money in the game, they borrowed money to invest in themselves! They were doing the same things they had done at Salomon but they failed to appreciate the importance of being able to lean on Solly's balance sheet in times of trouble.
As for Taleb ... puh-lease. The guy is a self-promoting windbag and his two most recent books are a waste of time. That's a shame, because he has written at least one interesting book I am aware of: Dynamic Hedging. Unfortuntely, the flaws that were present in embryo in that book - exaggerated self-regard, exaggerated criticism of others, deliberately cryptic statements meant to make the author seem clever - have grown like a tumour to consume 100% of his writing. All of Taleb's points have been made more clearly and more intelligently by other, better people. A recent example is Rebonato's Plight of the Fortune Tellers, but there are many others.
Re:Nothing wrong with models. (Score:5, Interesting)
Exactly. EVERY model that only sees rising house prices during it's data collection phase WILL assume that house prices will keep rising, and therefore tell bankers that dodgy mortgages are ok.
After all, as long as house prices keep rising, there is NO risk whatsoever in dodgy mortgages. Either you get the stated intrest (buyer pays mortgage) or you get the price rise of the house since the buyer bought it with your money (in the case of default) ... the risk of losing money in the deal is EXACTLY the chance that house prices drop. And house prices never dropped (significantly) in over 50 years ... obviously any statistical algorithm would have told you the risk was minimal.
Re:Picking up pennies in front of bulldozers (Score:1, Interesting)
That and the fact that you could insure a bond you didn't own. There were more credit default swaps "insuring" bond than actual bond. Also, the bonds were from the same pool of companies doing the insuring. A clear total house of cards which if regulated would have had the regulators screaming on capital hill long before it got that far.
Sure he did. He kept interest rates low which kept the bubble growing instead of popping. Then when it looked like the choice was serious stagflation or popping the bubble, he quit. I wouldn't call that nothing.
Re:Nothing wrong with models. (Score:4, Interesting)
>>>The LTC crash was caused by the founders (Nobel Laureates in Economics) having a model
Smart people always think they are so smart - until they discover that they too can make mistakes. I was just watching a video wherein a Cop explained how he outsmarted a lawyer - or more correctly the lawyer outsmarted himself. The lawyer was so absolutely certain that he knew the law & couldn't be caught, but then he bragged about it ("I know you can't trace sales in flea markets, because it's forbidden by law"), and off the cop went and bought the murder weapon. The lawyer was correct about the law, but overlooked the cop's willingness to do some old-fashioned footwork. Ooops.
No matter how smart you are, you can still make mistakes. Your real estate model had a mistake in it - as you just discovered with this market bust.
Re:Nothing wrong with models. (Score:5, Interesting)
Even more important that the limitations of a model are the assumtions taken in developing the model and/or feeding the data into the model, these should always be made clear to whomever the user of the model is, and it is then up to the user to decide if those assumtions are reasonable for their use of it.
The problem with this is most people's "just give me what I need to get the job done today" attitude. I've taught statistics in community college for a number of years, and I grapple with this a lot. Difficult enough to get people to perform the calculations for z-interval/test. Almost impossible to get them to consider the meta-analysis on whether the test is legitimate (simple random sample, assessment of normal population if sample size small, known standard deviation, etc.)
If most days they can get away with ignoring the model's assumptions, then folks wind up doing so, and then that knowledge degenerates. Ultimately the exceptional day that they need that skill, they don't have it. People function very, very poorly in relation to very infrequent (once a generation?), catastrophic events.
Re:Nothing wrong with models. (Score:2, Interesting)
"Bankers should have noted that very small changes in their underlying assumptions could result in very large changes in the correlation number."
Any mathematical, statistical, financial, etc. model has underlying assumptions built into the model. In academia, you are taught to check the assumptions of these models before you EVER consider using them as a part of the solution. In the real-world, this doesn't happen as often as it should. Agreed, nothing wrong with the models. Wrong application of the models.
Re:the formula that killed wall street: (Score:4, Interesting)
Those that work at companies that are entirely family or employ owned, do you feel that your company is in better shape than those public stock corporations?
Without a doubt, yes.
I own 3 companies - all three are growing through this recession. None had any debt.
Re:the formula that killed wall street: (Score:3, Interesting)
Well we have no debt and about 4 months worth of costs in savings, does that answer your question? :)
Re:Economic Stimulus (Score:4, Interesting)
You know I've been hearing this for years, so I actually looked it up. As far as I can see [treas.gov] China "only" has about $700 billion of US government debt. A huge amount certainly, but really enough to cause the kind of financial armageddon that people talk about?
Re:One word (Score:3, Interesting)
However, you could still be making money right now if you know what you are doing, but for most people, their 401(k) is kind of a "buy and ignore" type operation.
Any ideas or insight into how? There is only so many options offered in a typical 401(k). I don't see how you can say what you just said. Further explanation would be much appreciated.
Laying blame (Score:2, Interesting)
Slashdot mods broken--again (Score:3, Interesting)
Who modded this insightful?
Lenders set the interest rate to be higher than monetary expansion. If they didn't do this, they'd lose real value.
Money is credit. If the parent was correct, the money supply would be expanding at around a typical debt interest rate (say 5%). Yet there are many stable economies where this has not been the case for a long while--every economy that ever used gold, for instance. Moneylenders didn't conjure gold into existence by setting interest rates.
The answer is even simpler than that. Amazingly enough, you can pay off debt without increasing the supply in credit. Do Slashdot mods really think you have to magic a dollar bill (or, pre fiat money, a gold coin) out of thin air to pay interest? That interest may not be paid from existing credit or debt? And once it gets paid, the token of exchange disappears forever? Stupid!
Re:Citation, please (Score:3, Interesting)
I keep posting the link in my sig when I feel it is appropriate, and while I fear I'm beginning to sound like a broken record, now is one of those times.
The Crash Course [chrismartenson.com] is a 20-part video series, most chapters under 7 minutes long, that explores the various issues that are all coming to a head, including peak oil, world population, global warming, the money supply, and a few others. It's full of exponential curves that would be exciting except for the ceilings they are approaching at an increasingly alarming rate.
It's pretty well put together and highly informative. If you're skeptical about all the gloom and doom because it's usually spouted by ACs, you owe it to yourself to get the information from a source far less shrill and far more credible. IMO you'll be pretty convinced after watching, and even if you're not you will certainly be better informed.
Re:Nothing wrong with models. (Score:3, Interesting)
According to the all-knowing wikipedia [wikipedia.org], the government debt increased from $6e12 to $9e12 during Bush's term. However, this was during a time of huge growth in the GDP. Looking at the graph as a percentage of GDP, government debt remained fairly constant.
I'm not at all a fan of Bush's big government spending, he should have reduced spending while he had the chance. That being said, I find it ridiculously hypocritical for Bush pundits to complain about his government spending then but not Obama's drastic spending now. Bush may have increased the government's debt by 50% during his 8 years in office, but Obama has increased it by 50% in his first 60 days, AND during a huge decrease in the GDP due to the recession.
Re:Nothing wrong with models. (Score:3, Interesting)
peragrin, you are wrong.
The current delinquency rate is over 6% for residential real estate and skyrocketing with no apparent end in sight. http://www.calculatedriskblog.com/2009/02/fed-delinquency-rates-rise-sharply-in.html [calculatedriskblog.com]
This is relatively new data, but I have been following the mortgage markets closely since the credit crisis began as ARM/and Option ARM mortgages reset and the situation is certainly going to deteriorate a great deal.
Re:All models have limitations (Score:3, Interesting)
>>Say it with me: "All models are wrong. Some models are useful"
>>Newtonian physics is technically a less accurate model than Einstein's general relativity
Not in the world of economic models that do forecasting. Models of the physical world in various scientific fields are fine. Models that claim they can predict the future or quantify risk are dangerous. As in, they can destroy billions of dollars of wealth and potentially take out an entire nation's economy with it. Would the current downturn be as serious if we didn't have mortgage backed securities that had "quantified risk"? When people bought into the notion that these securities were low risk, it endangered the entire world economy. Iceland's bank has gone bankrupt.
>>They did not adequately consider the risks they were taking.
They *did* consider the risks they were taking. That's the point - the founders of LTCM had Nobel Laureates on this very subject! Everyone involved in the disaster thought that they had risk tamed by a mathematical model. Then they learned that reality trumps math.
It's especially bad when you model things using a Gaussian: LTCM used a Gaussian model for price movements. Essentially, by their model, large price movements (like what killed them) were near-impossible. IMO, Gaussians should only be used when you're looking at processes that really are the sum of large numbers of random events (http://en.wikipedia.org/wiki/Central_limit_theorem). It's grossly inappropriate to use them when it's not. Sure, short-term fluctuations in price might actually be adequately modeled by a Gaussian, but just one real life event throwing a monkeywrench into your system, and all of your little arbitrage gains are wiped out overnight.