How To Profit From Planetary-Scale Computing 178
An anonymous reader writes "MIT physicist Alex Wissner-Gross and mathematician Cameron Freer have devised a technique for exploiting geographic location in high-frequency trading, reports FastCompany. From the article: 'We view this work as one of the first serious, credible justifications for covering the planet's surface with computers. [...] We've perhaps identified a new type of natural resources that sovereignties might take advantage of.' Physicist and hedge-fund manager Jean-Philippe Bouchaud says, 'This shows that the technological arms race to extract every penny from high-frequency mechanical arbitrage will soon reach its ultimate limits.'"
Re:Why the snow (Score:5, Informative)
The traveller who has never before experienced an arctic summer, and who has been accustomed to think of Siberia as a land of eternal snow and ice, cannot help being astonished at the sudden and wonderful development of animal and vegetable life throughout that country in the month of June, and the rapidity of the transition from winter to summer in the course of a few short weeks. In the early part of June it is frequently possible to travel in 'the vicinity of Gizhiga upon dog-sledges, while by the last of the same month the trees are all in full leaf, primroses, cowslips, buttercups, valerian, cinquefoil, and labrador tea, blossom everywhere upon the higher plains and river banks, and the thermometer at noon frequently reaches 70 deg. Fahr. in the shade. There is no spring, in the usual acceptation of the word, at all. The disappearance of snow and the appearance of vegetation are almost simultaneous; and although the tundras or moss steppes, continue for some time to hold water like a saturated sponge, they are covered with flowers and blossoming blueberry bushes, and show no traces of the long, cold winter which has so recently ended.
George Kennan, Tent Life in Siberia [gutenberg.org]
The last time a planetary scale computer was built (Score:2, Informative)
Re:Time for a rant... (Score:5, Informative)
Hear hear.
I pulled out 100K out of the markets because I can't just put up with HFT anymore. So Buy and Hold was a bad idea. Now investing is a bad idea.
You can't put in a stop-loss order anymore on anything you own because every day you have to worry if a mini-flash crash hit one of your issues and triggered it, then the SX won't unwind YOUR trade but they are glad to unwind the fuck-up trades the HFT guys caused.
The utility of arbitrage (Score:3, Informative)
The thing is, arbitrage doesn't create liquidity, it simply capitalizes on the mistakes other people make.
Generally speaking arbitrage depends on the existence of liquidity (the ability to sell an asset without greatly moving the price) in order to work. It's impossible to capitalize on a "mis-priced" asset if there is no market for that asset. That doesn't mean however that arbitrage is without value. Price convergence [wikipedia.org] is a common result of arbitrage and it tends to reduce price discrimination.
In a certain sense, all business is an exercise in statistical arbitrage - exploiting the difference in prices between two or more markets. You buy goods where they are cheap (possibly assembling them) and sell them where they are dear. Without the ability to exploit price spreads profit is impossible. If someone makes a "mistake" in pricing, we should expect someone to step in to take advantage of that mistake.
Re:The utility of arbitrage (Score:3, Informative)
Businesses in your example introduce value, by taking care of shipping the products to/from the place of production to the place of sale.
Businesses DO introduce value by by shipping products where they are needed. They are exploiting the difference is value between two markets by doing so. That is by definition arbitrage.
It's not statistical arbitrage at all, which alone creates no value, only number games.
Sure it is - you just have to think about it in a slightly larger context. Statistical arbitrage occurs whenever there is a mispricing of price relationships that are true in expectation. In other words, you produce a good or service and ship it to market because you have an expectation of exploiting a difference in pricing. In the short term (just as with statistical arbitrage) events can occur that can introduce heavy short term losses. You also have to allow for the fact that people are imperfect and most decisions are made with imperfect information.
Oh, and arbitrage (statistical or otherwise) DOES often create value. It forces price convergence which in turn reduces price discrimination. This isn't always worth the costs but arbitrage does demonstrably have real value in the real world.
There is no economic point in buying/selling a product in a factory without it ever leaving the factory.
True but irrelevant to my point.
Re:Another way to look at this: (Score:1, Informative)
To what extent is so called high speed trading actually turning into electronic non-shooting warfare? Some of the techniques described are essentially variations on DDOS and spam. The recent Scandinavian case throws into question the point at which the techniques shade into illegality - is it just that if you or I do it, it is illegal, whereas if a bank does it, it's business as usual?
And to what extent is this latest proposal, while apparently to do with the distance between exchanges, also actually about putting resources into jurisdictions which have perhaps more elastic definitions of what constitutes legal trading?
On previous form, this will probably get moderated troll or flamebait. But it's actually two questions that I have never had adequately answered, except for the usual "you wouldn't understand" from the traders. If I, a graduate systems developer with further education in economics, can't understand them, what's the betting that our elected representatives can?
I'll try to answer your questions in the first paragraph. "Non-shooting warfare" is probably a good way to describe trading, whether high speed or not. Just like chess or Starcraft is non-shooting warfare. And also, unlike warfare, there's a strong mathematical argument that more trading is good for the market: more trading means more information being communicated, and reaching equilibrium more quickly. (In fact, many very smart people argue insider trading should be legal. I'm not sure I fully agree, but the fact is, the commodity market has loose insider trading laws and it seems to function very well. I wish I had a reference to some papers, but couldn't pull any up quickly.) Trading is most analagous to warfare in that you act purely in your own self interest: you buy when you think something is underpriced, and sell when it's overpriced, and--should your opponents be irrational--perhaps making trades to trick your opponents. The latter can fall in a gray area, since market manipulation is illegal, though the case law isn't entirely straightforward. Fortunately, markets with lots of traders and liquidity are the hardest to manipulate.
Legally there's no difference between what a bank and an individual trader can do. Actually the bank may have more pressure not to do sketchy trading strategies because (don't laugh, it's true) banks take their reputations very seriously, at least by the standards of the industry. Hedge funds, on the other hand, may be more willing to do gray-area trading since they do not have clients like banks do. I heard about the Scandinavian case, but I don't know all the details. Case law is a bit tricky as to what is market manipulation and what isn't. Certainly having good lawyers always helps, but they were definitely in a gray area.
Re:Limits? (Score:1, Informative)
You know how you can avoid ever having a huge national housing market crash? Easy. Limit the purchasing of non-commercial residential homes to people who actually intend to live in them. Do that and DON'T make "securities" out of them. Then you can't have a bubble in the first place -- no bubble, no burst.
What do you do with all the people renting investment properties now because they are not in a position to buy a home? If it wasn't for people investing in property there would be no rental market. I agree with you in principle but adjusting one thing has an effect on others. Solving the next problem then becomes the issue.
Real simple. If I own a house that I rent out and have no intention of living in, that would be a commercial residential home.
That's still absolutely nothing like taking numbers of such properties, dividing them up into shares, and selling them as securities. Dig?
Re:Limits? (Score:3, Informative)
Without an arbitrary investment in rental properties housing prices would fall. Supply and demand. It's a fallacy to think that if you flood the market with homes prices will drop never mind the current bust scenario. It is the land that has value and by turning land into investment properties you make land more scarce for those who would buy a home to live in. This drives up prices for the land itself. Without rental properties the developers would be out as they need those investors to buy up the surplus lots. We would return to buying individual lots, hiring a contractor and architect and having a home built.
That would be a good thing, employing many more skilled craftsman and less premanufactured homes built in factories.
Re:HFT is Not a Sin (Score:2, Informative)
I'm puzzled by your inference that humans already know the performance of a future stock. They don't. Nobody knows what's going to happen in the markets. All we have is educated guesses, and entrusting a computer to scalp trade for you is one possible educated guess to make.
Yes the events of the flash crash were alarming. But the stock market is a complex adaptive system. There have been new laws put in place, and if need be there will be more.
And as for why companies may be trading above their 120 day moving averages; perhaps it has something to do with the rebound from after the stocks crashed? After the market lost about 50% of it's value seem to me that any bounce back from that low would be a sharp one. The market has to come to an exaggerated fall to find the bottom.
Re:Time for a rant... (Score:2, Informative)
Lots of us just are refusing to play that game at this point.
It's sad, when I can go on a microfinance site, give money to some random, nearly unverified person across the internet, and not only have better confidence that my money is being well-kept, taken care of and safe, but also providing a better return than putting that money into the stock market, even with the most conservative of investing strategies!
Re:Time for a rant... (Score:3, Informative)
I presume you're talking about Kiva or something very similar. If so, I agree completely. I've lent my initial stake about 10 times, and each time it's an actual investment. I picked companies that I though could use effectively use the money, sent them the cash, then watched as they repaid it. It's a gratifying feeling, I tell you.