Monday, October 03, 2011 2:36:39 AM
Frankenstein finance: How supercomputers preying on human fear are taking over the world's stock markets
73 per cent of shares on the New York Stock Exchange are traded by computer
By Robert Harris
Last updated at 6:31 PM on 2nd October 2011
A spectre is haunting Europe: the spectre of capitalism. A vast and highly unstable mixture of debt — trillions of dollars of sovereign, corporate and private borrowing accumulated over decades — is strapped to the advanced Western economies like a suicide bomber’s gelignite vest.
The task facing our politicians is somehow to defuse this bomb without inadvertently triggering the sequence of defaults and bankruptcies that would set it off. No wonder they walk around the problem scratching their heads, prodding it gingerly here and there. The horrible truth is dawning that the problem may well not be technically solvable.
For the first time in my life — I am 54 — I get the sense of what it must have been like to have lived in my grandparents’ or great-grandparents’ generation: in 1913, say, or 1937. One feels a great smash coming ever closer, almost in slow-motion, and yet there seems to be nothing that can be done to avoid it.
How have we got ourselves into this mess? After all, we were supposed to be living in an era of unprecedented peace and prosperity.
Communism had collapsed and the threat of nuclear annihilation had receded. Immense advances in computer technology were creating whole new economies. Vast markets were opening up in the developing world. Above all, we were supposed to have learned enough about economics to have created the necessary institutions — the World Bank, the International Monetary Fund, the G20, the OECD — to ensure we never repeated the mistakes of the Thirties.
Where did it all go wrong?
Bizarrely enough, as good a place as any to start looking, in my opinion, is a hole in the ground near the small town of Waxahachie, Texas — or, to be more precise, 17 holes in the ground, each of them an air shaft leading down to 14 miles of abandoned tunnel dug into the hard Texan rock, which are all that remains of a grandiose scientific project called the Desertron.
The Desertron — or the superconducting super-collider, to give it its proper scientific title — was supposed to be America’s answer to CERN’s Large Hadron Collider in Geneva, a gigantic experiment to investigate the most fundamental laws of our universe. With a circumference of 54 miles, it would have been three times as large and powerful.
Unfortunately it would also have been nearly three times as expensive. In October 1993, in order to save projected future costs of $10 billion, the U.S. Congress voted to abandon the whole scheme — writing-off the work already done at a cost of $2 billion.
For a whole generation of American academic physicists, that decision wiped out their planned careers.
One physicist with a PhD I spoke to when I was researching my new novel, now in his 40s, told me he cried when he heard the news. What was he supposed to do now? He had to earn a living somewhere. His solution, like that of a majority of his colleagues, was to go and work on Wall Street — in his case, in the giant investment bank Merrill Lynch.
The resulting collision of brilliant but unworldly scientists trained to manipulate sub-atomic particles and aggressive financial traders eager to devise new products was to be more spectacularly dangerous than anything that might have been produced beneath the dusty surface of Waxahachie.
For this was the deadly partnership that helped give us a whole alphabet soup of fearsomely complicated financial derivatives — loans and mortgages and investments packaged up into bundles and sold around the world — that almost no one, and certainly not the regulatory authorities, ever really understood.
Only a matter of time: Billionaire investor Warren Buffet has criticised what he calls 'financial weapons of mass destruction'
When these toxic ‘financial weapons of mass destruction,’ as the U.S. billionaire Warren Buffett presciently called them, duly blew up in 2008, the same U.S. Congress that had saved $10 billion shutting down the Desertron had to come up with a rescue package for the banking system that has since been estimated as costing the American taxpayer $3.7 trillion.
If ever there was an example of the Law of Unintended Consequences in action, this must surely be it. But that may be only the start.
Before I began my research, I subscribed to the widely-held view that people in the financial sector generally had qualifications in economics or business, wore striped shirts and braces (if they were male), and sat in trading rooms shouting wildly into four phones simultaneously.
To my surprise, I found that this image is entirely outdated.
'Quant': Traders these days are more likely to have qualifications in maths or physics than economics
One extremely successful hedge fund manager I spoke to — with $12 billion in assets under management — won’t hire anyone without a top PhD in maths or physics; even economics is considered too ‘soft’ a degree. Increasingly, the people in the dealing rooms these days — young, casually-dressed — look as though they should be in lecture halls. They are known in the business as ‘quants’ — short for ‘quantitative analysts’.
Quants analyse the market with intense mathematical and statistical precision to predict share price movements and the level of investment risk; they sit at screens and rarely talk in anything louder than a whisper.
The trading is mostly done by computer, for which the quants write the programmes. Now, 73?per cent of shares in New York are traded by computer, either by so-called ‘high-frequency strategies’, which may hold the shares for only a few milliseconds, or by algorithms devised by quants. Algorithms are sophisticated programmes designed to predict the behaviour of the markets.
There is something slightly creepy about it. In the words of Emanuel Derman, himself a leading quant: ‘When physicists pursue the laws of the universe, it seems selfless. But watching quants pursue sacred laws for the profane production of profit, I sometimes find myself thinking disturbingly of worshippers at a black mass.’
Increasingly, the role of the trader is like that of a pilot in a fly-by-wire jumbo jet. The job is done by computers: he — for some reason quants are mostly men — sits at a screen and monitors the operation, only intervening when something goes wrong.
Recently I watched an algorithmic system in Geneva belonging to a hedge fund trading on the New York Stock Exchange. The computer had picked the stocks it wanted to trade. It communicated with the broker’s computerised system in the U.S. which, in turn, communicated with the computerised exchange that facilitated the deal. At no point was a human involved.
In the 20 minutes I was watching, the machine made a profit of $1.5 million. This hedge fund has made a return for its investors of more than 80?per cent in the past three years, at a time when most of us have seen the value of our pensions and tracker-funds go down in a falling market.
‘Our computers love it when the markets panic, because when people panic they behave in very predictable ways,’ I was told. In other words, the machines thrive on fear.
Volatility: Computerised trading thrives on fear
There is even a way of estimating this human weakness: the Standard & Poor 500 Volatility Index (VIX) measures the expected volatility on the Chicago Stock Exchange over the coming month, based on a hugely complicated mathematical formula devised by quants. It is popularly known as ‘The Fear Index’.
In 1965, the founder of the computer firm Intel, Gordon Moore, propounded what is known as Moore’s Law: that computers would double in power and halve in cost every 18 months. His forecast has proved amazingly accurate.
To take one example: as recently as the Nineties, CERN’s experimental data was all analysed by a Cray X-MP/48 supercomputer which cost the scientists $15 million. Yet that machine had less than half the computing power of a modern Microsoft Xbox, which costs $200.
When something continues to double in size — in this case computer power — it is called exponential growth. But as Moore himself observed a few years ago, exponential growth can’t continue for ever. It can be pushed to its limit, he said, ‘but eventually disaster happens’.
We have been warned.
Computers have become so powerful in the world of financial trading that the human involvement has been reduced to that of the quants and their obsessive statistical analyses. But computer programmes based on statistics, however brilliantly analysed, do not allow for common sense.
Computers predicted the U.S. property market would rise for ever because statistics showed the country’s house prices had never fallen in history — and every financial institution worldwide piled into the American’s mortgage market. We all know what happened next: the housing market crashed, U.S. mortgages were worthless and the so-called sub-prime loan crisis sent the world’s financial markets into meltdown.
Most of us in Britain remember May 6 last year as the date of the General Election. But about two and a half hours before the polls closed in the UK, at around 2:30 pm on the American East Coast, the U.S. financial markets experienced what came to be known as ‘the Flash Crash’.
The events of those few minutes provide a terrifying snapshot of what the modern markets have become. First, there is their sheer scale: 19.4 billion shares were traded on that day, more than were traded in the entirety of the Sixties.
But the figure is misleading: hundreds of millions of these shares were never actually sold, but merely held for a few thousandths of a second as computerised high-frequency traders tested the waters in the market.
They ‘sniped’ and ‘sniffed’ (in the jargon of the industry), making bogus offers to buy or sell shares so that they could find out the price, but the traders never went through with the sale.
The trouble is that the computers registered these bogus offers as real sales, and so much of this activity took place that the online trading section of the New York Stock Exchange temporarily froze. It was unable to cope — all the ‘sniping’ and ‘sniffing’ had made the amount of shares traded seem ten times larger than it actually was.
Drop: Huge lurches in the markets are now common
In the ensuing panic, the Dow Jones Industrial index dropped by roughly 700 points in the space of 20 minutes, wiping out nearly $1 trillion of investors’ money. This, then, is the financial world which we now live in: a world of extreme volatility, with lurches of 3 or 4?per cent a day on the markets no longer uncommon.
A world of terrifyingly complicated financial instruments designed to spread risk but which have, instead, spread a contagious lack of confidence; a world of instant communications, in which tremors of panic spread across the whole globe in the time it takes ripples to spread across a lake; a world in which thousands of the most brilliant minds on the planet are no longer paid to pursue scientific progress, but to devise financial strategies that are mostly non-productive and sometimes potentially highly dangerous.
The novelist and physicist C.?P. Snow delivered a famous lecture in 1959 about what he called ‘the two cultures’, the humanities and the sciences, and the failure of the one to understand the other.
‘A good many times,’ he said, ‘I have been present at gatherings of people who, by the standards of the traditional culture, are thought highly educated and who have with considerable gusto been expressing their incredulity at the illiteracy of scientists.
‘Once or twice I have asked the company how many of them could describe the Second Law of Thermodynamics. The response was cold: it was also negative. Yet I was asking something which is the scientific equivalent of: “Have you read a work of Shakespeare’s?” ’
It seems to me we now have to add a third culture to Snow’s list: the financial markets.
How many of us, for example, have the least idea of what the latest financial instrument — an exchange-traded fund — actually is? Yet the quants are now busy turning exchange-traded funds into a trillion-dollar industry, up 40 per cent in Europe alone in the past year. And like all these computer-devised financial derivatives that preceded them, they are cloaked in mystery and are risk damaging the financial system.
How many of us even know what short-selling is? I certainly didn’t, before I started researching my book.
As the current sense of sleepwalking towards calamity continues, my worry therefore is not so much the obviously imminent Greek default, or even the strains in the Eurozone, or the U.S. budget deficit, or the long-term intentions of the Chinese.
It is that the financial system itself has somehow slipped all human control — that it has become the preserve of a profoundly anti-democratic, super-rich elite, and that it girdles the planet like some alien entity from an H.?G. Wells novel.
The digitised financial machine does not work for us: we work for the machine. And I do not believe that our political leaders have the faintest idea how to bring it under control.
Robert Harris’s new novel, The Fear Index, published by Hutchinson, is out now.
© Associated Newspapers Ltd
http://www.dailymail.co.uk/debate/article-2043943/How-supercomputers-preying-human-fear-taking-worlds-stock-markets.html [with comments]
===
Light is not fast enough for high-speed stock trading
Stripping back automated computer trading could shave milliseconds off trades – perfect for traders out to make their millions
01 October 2011 by Jeff Hecht
Magazine issue 2832
EVERY microsecond counts in stock trading [ http://www.newscientist.com/article/mg20827855.500-automated-traders-eye-up-unlikely-locations.html ]. The New York Stock Exchange handles a third of the world's stock trading - around 22 billion messages a day. But NYSE Euronext [ http://corporate.nyx.com/en/home ], which operates the exchange, wants it to get even faster.
Now cable company Hibernia Atlantic [ http://www.hiberniagfn.com/index.php ] is spending $300 million to build a new transatlantic cable to shave 6 milliseconds from the present 65-millisecond transit time between London and New York. It will be the first new cable to cross the Atlantic in a decade and trading firms are likely to pay premium rates to use it.
This is because even though a computer can execute millions of instructions in a microsecond, the furthest light can travel in that time - even in a vacuum - is just 330 metres. That is an age if algorithms are competing to execute the best trades.
"The speed-of-light limitation [ http://www.newscientist.com/article/dn20957-dimensionhop-may-allow-neutrinos-to-cheat-light-speed.html ] is getting annoying," Andrew Bach, head of network services at NYSE Euronext, told the European Conference on Optical Communications in Geneva, Switzerland, last week.
With global markets currently in turmoil [ http://www.bbc.co.uk/news/business-15031731 ], it might seem a strange time to worry about the speed of trades, particularly when automated trading was implicated in the stock market's May 2010 "flash crash" [ http://money.cnn.com/2010/10/01/markets/SEC_CFTC_flash_crash/index.htm ]. But traders still want their computers to receive trading data and place orders instantaneously [ http://www.newscientist.com/blogs/onepercent/2011/07/high-speed-trading-algorithms.html ]. And customers will go elsewhere if a rival is faster.
Bach proposes speeding up signals by shifting from the solid-core optical fibres used at present to hollow-core fibres. Glass slows light down by about a third, says Philip Russell, who studies hollow-core fibres at the Max Planck Institute for the Science of Light in Erlangen, Germany.
Bach also suggests a few changes to the way in which signals are sent. One is to stop data compression, normally done to save bandwidth. "We can't afford 2 to 3 microseconds to compress data," he says. This would mean less information could be sent at once, but it would be quicker.
A second is to abandon error-correction codes, which improve signal quality but take time to process, by designing a system that is more error-resistant. Ending the practice of retransmitting lost data packets could also help, says Bach. This can confuse computer-trading software by scrambling the sequence of trades. Even random fluctuations of 1 to 2 microseconds in the arrival of data packets, an effect called jitter, can confuse computer algorithms.
Some observers worry that further stepping up the pace may be a bad idea. In July, Andrew Haldane of the Bank of England warned [ http://www.newscientist.com/blogs/onepercent/2011/07/high-speed-trading-algorithms.html ] that "flash crashes, like car crashes, may be more severe the greater the velocity".
In contrast, Terrence Henderschott of the University of California, Berkeley, finds "no compelling evidence" for automated trading causing problems.
© Copyright Reed Business Information Ltd.
http://www.newscientist.com/article/mg21128324.700-light-is-not-fast-enough-for-highspeed-stock-trading.html [with comments]
===
(linked in):
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67191187 and preceding and following
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=66175494 and preceding and following
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67609623 and preceding (and any future following)
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67470083 and preceding and following
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67528249 and following
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67612684 and following
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67556425 and preceding (and any future following)
73 per cent of shares on the New York Stock Exchange are traded by computer
By Robert Harris
Last updated at 6:31 PM on 2nd October 2011
A spectre is haunting Europe: the spectre of capitalism. A vast and highly unstable mixture of debt — trillions of dollars of sovereign, corporate and private borrowing accumulated over decades — is strapped to the advanced Western economies like a suicide bomber’s gelignite vest.
The task facing our politicians is somehow to defuse this bomb without inadvertently triggering the sequence of defaults and bankruptcies that would set it off. No wonder they walk around the problem scratching their heads, prodding it gingerly here and there. The horrible truth is dawning that the problem may well not be technically solvable.
For the first time in my life — I am 54 — I get the sense of what it must have been like to have lived in my grandparents’ or great-grandparents’ generation: in 1913, say, or 1937. One feels a great smash coming ever closer, almost in slow-motion, and yet there seems to be nothing that can be done to avoid it.
How have we got ourselves into this mess? After all, we were supposed to be living in an era of unprecedented peace and prosperity.
Communism had collapsed and the threat of nuclear annihilation had receded. Immense advances in computer technology were creating whole new economies. Vast markets were opening up in the developing world. Above all, we were supposed to have learned enough about economics to have created the necessary institutions — the World Bank, the International Monetary Fund, the G20, the OECD — to ensure we never repeated the mistakes of the Thirties.
Where did it all go wrong?
Bizarrely enough, as good a place as any to start looking, in my opinion, is a hole in the ground near the small town of Waxahachie, Texas — or, to be more precise, 17 holes in the ground, each of them an air shaft leading down to 14 miles of abandoned tunnel dug into the hard Texan rock, which are all that remains of a grandiose scientific project called the Desertron.
The Desertron — or the superconducting super-collider, to give it its proper scientific title — was supposed to be America’s answer to CERN’s Large Hadron Collider in Geneva, a gigantic experiment to investigate the most fundamental laws of our universe. With a circumference of 54 miles, it would have been three times as large and powerful.
Unfortunately it would also have been nearly three times as expensive. In October 1993, in order to save projected future costs of $10 billion, the U.S. Congress voted to abandon the whole scheme — writing-off the work already done at a cost of $2 billion.
For a whole generation of American academic physicists, that decision wiped out their planned careers.
One physicist with a PhD I spoke to when I was researching my new novel, now in his 40s, told me he cried when he heard the news. What was he supposed to do now? He had to earn a living somewhere. His solution, like that of a majority of his colleagues, was to go and work on Wall Street — in his case, in the giant investment bank Merrill Lynch.
The resulting collision of brilliant but unworldly scientists trained to manipulate sub-atomic particles and aggressive financial traders eager to devise new products was to be more spectacularly dangerous than anything that might have been produced beneath the dusty surface of Waxahachie.
For this was the deadly partnership that helped give us a whole alphabet soup of fearsomely complicated financial derivatives — loans and mortgages and investments packaged up into bundles and sold around the world — that almost no one, and certainly not the regulatory authorities, ever really understood.
Only a matter of time: Billionaire investor Warren Buffet has criticised what he calls 'financial weapons of mass destruction'
When these toxic ‘financial weapons of mass destruction,’ as the U.S. billionaire Warren Buffett presciently called them, duly blew up in 2008, the same U.S. Congress that had saved $10 billion shutting down the Desertron had to come up with a rescue package for the banking system that has since been estimated as costing the American taxpayer $3.7 trillion.
If ever there was an example of the Law of Unintended Consequences in action, this must surely be it. But that may be only the start.
Before I began my research, I subscribed to the widely-held view that people in the financial sector generally had qualifications in economics or business, wore striped shirts and braces (if they were male), and sat in trading rooms shouting wildly into four phones simultaneously.
To my surprise, I found that this image is entirely outdated.
'Quant': Traders these days are more likely to have qualifications in maths or physics than economics
One extremely successful hedge fund manager I spoke to — with $12 billion in assets under management — won’t hire anyone without a top PhD in maths or physics; even economics is considered too ‘soft’ a degree. Increasingly, the people in the dealing rooms these days — young, casually-dressed — look as though they should be in lecture halls. They are known in the business as ‘quants’ — short for ‘quantitative analysts’.
Quants analyse the market with intense mathematical and statistical precision to predict share price movements and the level of investment risk; they sit at screens and rarely talk in anything louder than a whisper.
The trading is mostly done by computer, for which the quants write the programmes. Now, 73?per cent of shares in New York are traded by computer, either by so-called ‘high-frequency strategies’, which may hold the shares for only a few milliseconds, or by algorithms devised by quants. Algorithms are sophisticated programmes designed to predict the behaviour of the markets.
There is something slightly creepy about it. In the words of Emanuel Derman, himself a leading quant: ‘When physicists pursue the laws of the universe, it seems selfless. But watching quants pursue sacred laws for the profane production of profit, I sometimes find myself thinking disturbingly of worshippers at a black mass.’
Increasingly, the role of the trader is like that of a pilot in a fly-by-wire jumbo jet. The job is done by computers: he — for some reason quants are mostly men — sits at a screen and monitors the operation, only intervening when something goes wrong.
Recently I watched an algorithmic system in Geneva belonging to a hedge fund trading on the New York Stock Exchange. The computer had picked the stocks it wanted to trade. It communicated with the broker’s computerised system in the U.S. which, in turn, communicated with the computerised exchange that facilitated the deal. At no point was a human involved.
In the 20 minutes I was watching, the machine made a profit of $1.5 million. This hedge fund has made a return for its investors of more than 80?per cent in the past three years, at a time when most of us have seen the value of our pensions and tracker-funds go down in a falling market.
‘Our computers love it when the markets panic, because when people panic they behave in very predictable ways,’ I was told. In other words, the machines thrive on fear.
Volatility: Computerised trading thrives on fear
There is even a way of estimating this human weakness: the Standard & Poor 500 Volatility Index (VIX) measures the expected volatility on the Chicago Stock Exchange over the coming month, based on a hugely complicated mathematical formula devised by quants. It is popularly known as ‘The Fear Index’.
In 1965, the founder of the computer firm Intel, Gordon Moore, propounded what is known as Moore’s Law: that computers would double in power and halve in cost every 18 months. His forecast has proved amazingly accurate.
To take one example: as recently as the Nineties, CERN’s experimental data was all analysed by a Cray X-MP/48 supercomputer which cost the scientists $15 million. Yet that machine had less than half the computing power of a modern Microsoft Xbox, which costs $200.
When something continues to double in size — in this case computer power — it is called exponential growth. But as Moore himself observed a few years ago, exponential growth can’t continue for ever. It can be pushed to its limit, he said, ‘but eventually disaster happens’.
We have been warned.
Computers have become so powerful in the world of financial trading that the human involvement has been reduced to that of the quants and their obsessive statistical analyses. But computer programmes based on statistics, however brilliantly analysed, do not allow for common sense.
Computers predicted the U.S. property market would rise for ever because statistics showed the country’s house prices had never fallen in history — and every financial institution worldwide piled into the American’s mortgage market. We all know what happened next: the housing market crashed, U.S. mortgages were worthless and the so-called sub-prime loan crisis sent the world’s financial markets into meltdown.
Most of us in Britain remember May 6 last year as the date of the General Election. But about two and a half hours before the polls closed in the UK, at around 2:30 pm on the American East Coast, the U.S. financial markets experienced what came to be known as ‘the Flash Crash’.
The events of those few minutes provide a terrifying snapshot of what the modern markets have become. First, there is their sheer scale: 19.4 billion shares were traded on that day, more than were traded in the entirety of the Sixties.
But the figure is misleading: hundreds of millions of these shares were never actually sold, but merely held for a few thousandths of a second as computerised high-frequency traders tested the waters in the market.
They ‘sniped’ and ‘sniffed’ (in the jargon of the industry), making bogus offers to buy or sell shares so that they could find out the price, but the traders never went through with the sale.
The trouble is that the computers registered these bogus offers as real sales, and so much of this activity took place that the online trading section of the New York Stock Exchange temporarily froze. It was unable to cope — all the ‘sniping’ and ‘sniffing’ had made the amount of shares traded seem ten times larger than it actually was.
Drop: Huge lurches in the markets are now common
In the ensuing panic, the Dow Jones Industrial index dropped by roughly 700 points in the space of 20 minutes, wiping out nearly $1 trillion of investors’ money. This, then, is the financial world which we now live in: a world of extreme volatility, with lurches of 3 or 4?per cent a day on the markets no longer uncommon.
A world of terrifyingly complicated financial instruments designed to spread risk but which have, instead, spread a contagious lack of confidence; a world of instant communications, in which tremors of panic spread across the whole globe in the time it takes ripples to spread across a lake; a world in which thousands of the most brilliant minds on the planet are no longer paid to pursue scientific progress, but to devise financial strategies that are mostly non-productive and sometimes potentially highly dangerous.
The novelist and physicist C.?P. Snow delivered a famous lecture in 1959 about what he called ‘the two cultures’, the humanities and the sciences, and the failure of the one to understand the other.
‘A good many times,’ he said, ‘I have been present at gatherings of people who, by the standards of the traditional culture, are thought highly educated and who have with considerable gusto been expressing their incredulity at the illiteracy of scientists.
‘Once or twice I have asked the company how many of them could describe the Second Law of Thermodynamics. The response was cold: it was also negative. Yet I was asking something which is the scientific equivalent of: “Have you read a work of Shakespeare’s?” ’
It seems to me we now have to add a third culture to Snow’s list: the financial markets.
How many of us, for example, have the least idea of what the latest financial instrument — an exchange-traded fund — actually is? Yet the quants are now busy turning exchange-traded funds into a trillion-dollar industry, up 40 per cent in Europe alone in the past year. And like all these computer-devised financial derivatives that preceded them, they are cloaked in mystery and are risk damaging the financial system.
How many of us even know what short-selling is? I certainly didn’t, before I started researching my book.
As the current sense of sleepwalking towards calamity continues, my worry therefore is not so much the obviously imminent Greek default, or even the strains in the Eurozone, or the U.S. budget deficit, or the long-term intentions of the Chinese.
It is that the financial system itself has somehow slipped all human control — that it has become the preserve of a profoundly anti-democratic, super-rich elite, and that it girdles the planet like some alien entity from an H.?G. Wells novel.
The digitised financial machine does not work for us: we work for the machine. And I do not believe that our political leaders have the faintest idea how to bring it under control.
Robert Harris’s new novel, The Fear Index, published by Hutchinson, is out now.
© Associated Newspapers Ltd
http://www.dailymail.co.uk/debate/article-2043943/How-supercomputers-preying-human-fear-taking-worlds-stock-markets.html [with comments]
===
Light is not fast enough for high-speed stock trading
Stripping back automated computer trading could shave milliseconds off trades – perfect for traders out to make their millions
01 October 2011 by Jeff Hecht
Magazine issue 2832
EVERY microsecond counts in stock trading [ http://www.newscientist.com/article/mg20827855.500-automated-traders-eye-up-unlikely-locations.html ]. The New York Stock Exchange handles a third of the world's stock trading - around 22 billion messages a day. But NYSE Euronext [ http://corporate.nyx.com/en/home ], which operates the exchange, wants it to get even faster.
Now cable company Hibernia Atlantic [ http://www.hiberniagfn.com/index.php ] is spending $300 million to build a new transatlantic cable to shave 6 milliseconds from the present 65-millisecond transit time between London and New York. It will be the first new cable to cross the Atlantic in a decade and trading firms are likely to pay premium rates to use it.
This is because even though a computer can execute millions of instructions in a microsecond, the furthest light can travel in that time - even in a vacuum - is just 330 metres. That is an age if algorithms are competing to execute the best trades.
"The speed-of-light limitation [ http://www.newscientist.com/article/dn20957-dimensionhop-may-allow-neutrinos-to-cheat-light-speed.html ] is getting annoying," Andrew Bach, head of network services at NYSE Euronext, told the European Conference on Optical Communications in Geneva, Switzerland, last week.
With global markets currently in turmoil [ http://www.bbc.co.uk/news/business-15031731 ], it might seem a strange time to worry about the speed of trades, particularly when automated trading was implicated in the stock market's May 2010 "flash crash" [ http://money.cnn.com/2010/10/01/markets/SEC_CFTC_flash_crash/index.htm ]. But traders still want their computers to receive trading data and place orders instantaneously [ http://www.newscientist.com/blogs/onepercent/2011/07/high-speed-trading-algorithms.html ]. And customers will go elsewhere if a rival is faster.
Bach proposes speeding up signals by shifting from the solid-core optical fibres used at present to hollow-core fibres. Glass slows light down by about a third, says Philip Russell, who studies hollow-core fibres at the Max Planck Institute for the Science of Light in Erlangen, Germany.
Bach also suggests a few changes to the way in which signals are sent. One is to stop data compression, normally done to save bandwidth. "We can't afford 2 to 3 microseconds to compress data," he says. This would mean less information could be sent at once, but it would be quicker.
A second is to abandon error-correction codes, which improve signal quality but take time to process, by designing a system that is more error-resistant. Ending the practice of retransmitting lost data packets could also help, says Bach. This can confuse computer-trading software by scrambling the sequence of trades. Even random fluctuations of 1 to 2 microseconds in the arrival of data packets, an effect called jitter, can confuse computer algorithms.
Some observers worry that further stepping up the pace may be a bad idea. In July, Andrew Haldane of the Bank of England warned [ http://www.newscientist.com/blogs/onepercent/2011/07/high-speed-trading-algorithms.html ] that "flash crashes, like car crashes, may be more severe the greater the velocity".
In contrast, Terrence Henderschott of the University of California, Berkeley, finds "no compelling evidence" for automated trading causing problems.
© Copyright Reed Business Information Ltd.
http://www.newscientist.com/article/mg21128324.700-light-is-not-fast-enough-for-highspeed-stock-trading.html [with comments]
===
(linked in):
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67191187 and preceding and following
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=66175494 and preceding and following
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67609623 and preceding (and any future following)
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67470083 and preceding and following
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67528249 and following
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67612684 and following
http://investorshub.advfn.com/boards/read_msg.aspx?message_id=67556425 and preceding (and any future following)
Greensburg, KS - 5/4/07
"Eternal vigilance is the price of Liberty."
from John Philpot Curran, Speech
upon the Right of Election, 1790
F6
Join the InvestorsHub Community
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
