After the “ flash clangor ” that brought the stock rally to its knees a few months ago , information analytic thinking firm Nanex started look into some rum design in ancestry trade . They discovered bot armies are affecting deal on a microsecond - by - microsecond basis .
Over at The Atlantic , Alexis Madrigalhas a captivating storyabout how Nanex software engineer Jeffrey Donovan started analyzing market datum in a way nobody had ever done before . Instead of appear at business deal and quotes served up each second or minute , he dig deeper . He looked for market patterns at the sub - second level , and notice some seriously bizarre thing were fall out . Trading bots , control by obscure algorithms , were doing thing like requesting stock citation at a pace of 56,000 per second .
So are these algorithm create by malicious hackers , stress to repair the Flash Crash ? That is one possibility , since sometimes they bespeak so much data that human trading could be affected . What seems sealed is that they ’re not trace from the bot used by high - frequency traders , who often use algorithmic program to figure out where to buy and sell apace and constantly . These bots ’ entropy requests and trades are just too random to be useful for in high spirits - relative frequency dealer .
Madrigal writes :
The algorithms . . . do n’t process any procedure in the market . University of Pennsylvania finance professor , Michael Kearns , a specializer in algorithmic trading , anticipate the patterns “ curious , ” and note that it was n’t immediately apparent what such decree placement scheme might do .
Donovan think that the singular algorithmic rule are just a style of introducing dissonance into the piece of work . Other firms have to parcel out with that noise , but the originating entity can well strain it out because they lie with what they did . Perhaps that give way them an advantage of some milliseconds . In the highly free-enterprise and fast HFT world , where even one ’s physical proximity to a caudex exchange issue , market players could be looking for any advantage .
“ They are moving the eminent - frequence services as close to the exchanges as possible because even the speed of light matters , ” in such a free-enterprise marketplace , say Stanford finance professor Peter Hansen .
Donovan calls the visualizations of the data he found “ crop circle , ” andhe ’s created a web log where he post a daily chart of unexplained bot activity . He says he finds several examples of weird algorithmic behavior at the sub - 2d level every day , so obviously the market service computers are teem with bot body process .
The question stay : Who is hold these bot ? What on the button are they trying to do ? Some psychoanalyst dissolve these nanosecond form as mere chance , the emerging properties of a complex system . Others , like Donovan , are just trying to gather information and understand them . But I call up you and I know exactly what ’s go on . Out of the flyspeck pulse of a million bots , an artificial intelligence is being born .
Below , Madrigal offers us a few figure from Donovan ’s archive , with explanations . study his whole clause over atThe Atlantic . And check out the bot “ crop circles ” overat Donovan ’s Nanex blog too .
Madrigal ’s gloss :
This is an extreme closeup of just one endorsement of trading of the stock SHG , the Shinhan Financial Group . This is 760 quotes from a aggregate of 10,000 made in 12 seconds .
Here we see a “ flag repeater ” being executed on the BATS Exchange , the third - largest fairness securities industry after the NYSE and NASDAQ . 15,000 quote petition were made in 11 seconds in a reiterate pattern . Each iteration up the inverted comma a centime until $ 9.36 , and then the algorithm went down the same mode , a centime at a meter .
This chart shows a different kind of strategy . It represents 56,000 quotes in one second all at the same price ( the top chart ) but with the sizing of the guild increase by one ( i.e. 100 shares ) all the way up to 40,000 .
Finally , we see what Donovan calls the “ stubby triangle ” chart . It establish high quotes being made and then immediately followed by a stub order of $ 0.01 ( basically cancel in most contexts ) . The quote is then remade at a lower price and follow with another stub quote . This cycle happened at the charge per unit of 380 quote a second . [ This last description was clear up thanks to the kindness of writer Joe Flood . ]
FuturismsoftwareStock marketTechnology
Daily Newsletter
Get the good tech , science , and culture word in your inbox daily .
tidings from the future , delivered to your present .
You May Also Like
