Meet SGX’s new member - Crazy Eyes
Algos like this one will thrive when world’s fastest trading engine launches here on Aug 15
Remisier kings
The cousins who have taken over Peter Lim’s throne View PDF
Popular posts from this blog
Ex-bank owner Agus Anwar declared bankrupt
Prominent former bank owner Agus Anwar was yesterday declared bankrupt by the High Court.
Two ex-UOBKH staff charged with lying to MAS over due diligence reports on a Catalist aspirant
TWO former senior employees of UOB Kay Hian Private Limited (UOBKH) were charged on Wednesday for allegedly lying to the Monetary Authority of Singapore (MAS) in relation to reports on a then Catalist aspirant. Lan Kang Ming, 38, and Wee Toon Lee, 34, each face three charges of providing MAS with false information in October 2018 in relation to due diligence reports on an unidentified company applying to list on the Catalist board of the Singapore Exchange. MAS said in a media statement on Wednesday that it was performing an onsite inspection of UOBKH between June and August 2018, to assess the latter's controls, policies and procedures in relation to its role as an issue manager for Initial Public Offering (IPOs). During the examination, Lan and Wee were said to have provided different versions of a due diligence report relating to background checks on a company applying to be listed on the Catalist board of the Singapore Exchange. UOBKH had acted as the issue manager
Comments
Algos like this one will thrive when world’s fastest trading engine launches here on Aug 15
By Andy Mukherjee
08 August 2011
Let me introduce you to some of your new neighbours in Singapore: Mr. Bagman and Mrs Crazy Eyes, and their young ones, Master Street Lamps and Miss Enchanted Forest. You have not seen them yet, but you soon will. Or maybe not.
They are not Singaporeans, permanent residents or foreign talent. Actually, they are not even human beings. But just because they are algorithms, or rules computers use to make decisions, it does not make them any less real, or less worthy of your attention.
When the Singapore Exchange (SGX) launches the world’s fastest trading engine next Monday, these and many more such trading algorithms - or ‘algos’ as they are called in the investment community - will potentially have a home in the equity market here.
If the algos do come, be nice to them. They may just be lines of software code, but they are now responsible for about 70 per cent of all the buying and selling on the United States equity market.
Algos perform a few profitable functions. Some of these computer programs are designed to jump on to fleeting, split-second mis-pricing in markets that human traders are simply unable to detect or react to.
Algos also help to hide large buy or sell orders by breaking them down into small pieces and fooling other traders. Another set of computer programs scours the universe of small, scattered, seemingly random trades to unmask the presence of algorithms.
Bagman, Crazy Eyes, Street Lamps and Enchanted Forest are, for instance, some of the algorithmic patterns uncovered by Nanex, a trading data company in Chicago.
On May 6 last year, the world got a flavour of the punch algos pack when, in a matter of 30 minutes, the Dow Jones index had its biggest-ever intra-day fall, leading to instant evaporation of $1 trillion in market value.
Computers, it seems, had run amok.
Prices quickly returned to normal, but disquiet remains. Whether ‘high-frequency traders’ - machines trading with one another using algorithms - are making markets vulnerable to more such shocks in future is a question that has, since that fateful day, left regulators around the world more than a little uneasy.
Last month, Mr. Andrew Haldane, executive director of financial stability at the Bank of England, gave a speech in Beijing titled ‘The Race to Zero’. The title of his address referred to the scramble among exchanges around the world to allow a trade in virtually no time. We are already there - well, almost.
Nasdaq OMX has brought down the cycle time - or latency, as it is called in industry jargon - to 250 microseconds. That is less than one-sixtieth of one-sixtieth of a second, or about 1,400 times faster than the blink of an eye.
The SGX’s new trading engine, which also uses Nasdaq OMX’s Genium INET technology, promises to be even quicker with a latency of less than 90 microseconds.
Algos thrive on speed. Not only do they require a trading engine that can complete transactions quickly, but they also work best when the computers sit as close as possible to Internet data cables so they can shave off a few more microseconds from the entire trading process.
The SGX has already started offering this service, which it calls ‘co-location’, to more than 50 customers.
‘We are seeing quite a lot of interest in this business,’ said SGX chief executive Magnus Bocker last Tuesday.
About 29 per cent of the derivatives turnover on the SGX last year came from high-frequency traders. Extremely quick trading in securities will become possible with the new $250 million trading engine, though Mr. Bocker does not expect algos to become an overnight rage in the equity market here. ‘It will be a gradual move,’ he said. Results will be visible in a year.
But any decision to call off the race to zero has to be a globally coordinated one.
There is no way Singapore can escape panic if algos cause mayhem on Wall Street. So nothing is gained if the SGX - or the Monetary Authority of Singapore - were to unilaterally decide to keep very tight reins on high-frequency trading, refusing to let it dominate the marketplace.
That does not mean safeguards are not being put in place. The SGX plans to introduce circuit breakers that will curb extreme price movement, as well as pre-execution checks to prevent market manipulation by high-frequency traders.
In the final analysis, though, it will be wishful to imagine a future in which algos have been fully tamed.
As Mr. Kevin Slavin, co-founder of a video-game company that developed such hits as Drop 7 and Parking Wars, said in a recent TED talk, algorithms are so pervasive now - and not just in finance - that they are like a ‘co-evolutionary force’.
We will have to try to understand them just as we attempt to fathom the mysteries of nature. Bye bye, human traders; hello, Bagman.