Trading algoritmico intraday nei mercati valutari...

...è la specialità di Eleuthera Capital, un hedge fund il cui fondatore Philippe Bonnefoy è intervistato nel video (cortesia di http://www.opalesque.tv/hedge-fund-videos/ )  che vi propongo qui sotto. Il fondo Eleuthera Currency Fund investe nei principali cross-ratios con un CAGR-obiettivo del 15-25% con una volatilità annualizzata contenuta (10-15%). La maggior parte delle posizioni vengono mantenute in portafoglio per una durata compresa tra i 5 minuti e le 48 ore, troppo lunga per essere definita una strategia di HFT ma decisamente sufficientemente breve per distinguerla dalla maggior parte delle strategie macro o di carry trade. 

The fund uses a fully automated short-term systematic trading system. The system deploys two main computer models. One model trades in the direction of short-term market trends. The second model looks for mean reversion trading opportunities. Mean reversion trading opportunities often appear when markets become sharply oversold or overbought following the release of economic news, geo-political events or following central bank or governmental activity. This model attempts to profit form the likelihood that there will be an overreaction to “market moving” events and that these effects will have a transitory impact after which prices will return to their previous short-term range. The typical trade duration of all positions in the fund is between 15 minutes and 48 hours.

The computer models use a highly filtered screening methodology which analyses short term price action using technical analysis and pattern recognition techniques.

The trading system receives an electronic aggregated price feed from major banks and market counterparties through which it will execute transactions when all screening criteria are approved. 

Si scrive Tobin, si legge Robin ma è da Dustbin?

L'Economist appena uscito dedica un articolo abbastanza critico alle proposte di riforma dell'auditing e all'introduzione di una tassa sulle transazioni finanziarie.

Sullo stesso argomento potete anche leggere il commento di Buttonwood che allarga il discorso cercando di contestualizzare un po' le scelte che si presentano ai governi, anche se in una prospettiva abbastanza Londracentrica:

Let me start with the proposition that the financial sector has grown to dominate the Anglo-Saxon economies in ways that are unhealthy, particularly as the best and brightest have been lured to finance by high salaries. In my view, however, this outcome is the result of easy credit policies that fuelled a series of asset booms, accompanied by the gearing up of balance sheets and the willingness of central banks to rescue the markets whenever they faltered. The result for 25 years was a one-way bet on asset prices that the finance sector took advantage of. This was not a free market, but a rigged game.(...)

In the long run, moves to force banks to have higher capital will reduce the scope for leveraged returns, and thus reduce those big bonuses. Banks can go back to being rather dull utilities. But even this process is fraught with problems. The Bank of England's financial policy statement this week said that

banks should take any opportunity they had to strengthen their levels of capital and liquidity so as to increase their capacity to absorb flexibly any future shocks, without constraining lending to the wider economy. This could include raising long-term funding whenever possible and ensuring that discretionary distributions reflected any reduction in profits.

That is a tough combination to pull off, and so far lending has suffered. On the first part of the proposal, bank shares have underperformed this year, making it hard to raise new equity. The second part sounds good; cut back on bonuses rather than cut back on bank lending. But the banks are likely to say to one another; after you. If one bank cuts bonuses, and the others don't, the "good" bank will lose a lot of staff. Maybe this is something regulators have to decree, not just suggest.

So why not try a transactions tax instead? EU governments need the money, after all, and there is evidence that excessive trading can lead to volatility. Some will argue that the existence of stamp duty on UK equity transactions shows it can be done. But who pays the duty? Anyone who has ever bought shares in the UK will know that it is passed straight through to the retail investor or to pension funds, insurance companies and mutual funds (which are the aggregated savings of private investors). The duty has done nothing to slow the rapid rise in bankers' bonuses, which are generated elsewhere. Such taxes are just passed straight through to clients. Hedge funds often get round the duty, by trading in contracts for difference, a kind of leveraged bet on share prices that are exempt from the tax.

But what about all those derivative deals that are "socially useless"; wouldn't the tax catch them? It might if it were worldwide. But the bulk of European financial trading occurs in London, and the the bulk of that trading is conducted by non-UK firms. It would take little effort to switch the business to New York or Geneva or Singapore. Note that an impact assessment of the tax (which the EU commission undertook) says it willreduce GDP by 0.5%. Let Europe sign up for a financial transactions tax when the Republican party votes in favour of it (but don't hold your breath).

Indeed, this idea is so anti-London that the UK government is bound to veto it. Which makes one wonder why it was proposed; was it a smokescreen to divert the attention of EU voters from the mess that leaders are making of the debt crisis?

Qualcuno dei lettori mi sa indicare un link con una buona spiegazione di come sia aggirata la stamp duty tax dello 0.5% (!!) in vigore al LSE che gli investitori retail come me pagano ma che evidentemente non affligge i trading desk di banche, hedge funds, ecc? Davvero solo mediante i CFD? Peraltro in questo modo non si finisce con l'incoraggiare proprio quell'eccesso di leveraging che rende il sistema più instabile? Non ci sono esenzioni per chi dichiara (a torto o a ragione) di prendere posizioni sul sottostante per scopi di hedging su contratti derivati?

E' inutile e dannoso introdurre tasse sulle transazioni finanziarie se sono immediatamente e facilmente aggirabili (tranne naturalmente dal parco buoi).

Traders più veloci dei neutrini!

Mentre i governi europei, trainati dalla Francia e dalla Germania, insistono per l'introduzione di una qualche forma di Tobin Tax continua il dibattito sul trading ad alta frequenza. Negli ultimi due mesi l'alta volatilità ha certamente favorito l'HFT. Secondo il WSJ in agosto

High-frequency traders have roughly tripled their stock trades this month, estimates Tabb Group, a markets-research firm in New York. That has boosted their share of overall U.S. stock trading volume to about 65%, up from about 53% during the months before the August turmoil, according to the research firm.

That increase has come amid unusually heavy trading; on Aug. 8, as the Dow Jones Industrial Average sank 635 points, New York Stock Exchange composite volume saw its fourth-biggest trading day on record.

On that same day, high-frequency traders made record profits of about $60 million in U.S. stock markets alone, Tabb estimates. That doesn't sound like a huge amount in the context of global markets, but that daily profit-while likely not sustainable, analysts and traders say-would translate into an annual gain of about $15 billion.

Tabb estimates profits for other days that week ranged from about $40 million to $56 million. By comparison, it estimates that high-frequency traders made $7.2 billion in U.S. stock markets in 2009, the highest on record; the firm previously projected total profit of the high-frequency trading industry of less than $5 billion this year.

La polemica sul trading ad alta frequenza e sugli aspetti manipolativi di alcune strategie impiegate è molto accesa.

Nella sua ultima lettera settimanale John Mauldin cita un post recente di Nanex relativo ad alcune trades che sembrano essere state effettuate viaggiando nel tempo (nel futuro, ovviamente...):

Heads up, you Junior Rocket Man Kids (remember those days?). Physicists are doingamazing things. My son Trey and I got a private tour this summer of CERN, the great physics lab in Geneva. Very cool. But Wall Street is also legendary for the number of physicists it hires to work on high-frequency trading programs. Evidently, they have figured out how to get trades done 190 milliseconds in the future. Is the race on to see who can cross the one-day mark? What is the speed of light when compared to the speed of money?

"Nanex: On September 15, 2011, beginning at 12:48:54.600, there was a time warp in the trading of Yahoo! (YHOO) stock. HFT has reached speeds faster than the speed-of-light,allowing time travel into the future. Up to 190 milliseconds into the future, or 0.19 fantaseconds is the record so far. It all happened in just over one second of trading, the evidence buried under an avalanche of about 19,000 quotes and 3,000 individual trade executions. The facts of the matter are indisputable. Based on official UQDF/UTDF exchange timestamps, there is unmistakable proof that YHOO trades were executed on quotes that didn't exist until 190 milliseconds later!" (http://www.nanex.net/Research/fantaseconds/fantaseconds.html)

Going forward in time is cool, and the same day I got the above notice I read that the physicists at CERN and in Italy have found subatomic particles that move slightly faster than the speed of light, making it possible to travel back in time (only a few nanoseconds, but it's a start):

"But now it seems that researchers working in one of the world's largest physics laboratories, under a mountain in central Italy, have recorded particles travelling at a speed that is supposedly forbidden by Einstein's theory of special relativity.

"Scientists at the Gran Sasso facility will unveil evidence on Friday that raises the troubling possibility of a way to send information back in time, blurring the line between past and present and wreaking havoc with the fundamental principle of cause and effect.

"Researchers on the Opera (Oscillation Project with Emulsion-tRacking Apparatus) experiment recorded the arrival times of ghostly subatomic particles called neutrinos sent from Cern on a 730km journey through the Earth to the Gran Sasso lab."

( http://www.guardian.co.uk/science/2011/sep/22/faster-than-light-particlesneutrinos?newsfeed=true )

Now, just in case you buy this (and if you did, contact me about a bridge I have), let meattempt to disappoint. First, as my curmudgeon PhD from MIT and VC friend Bart Stuck writes, "I think they both had time-stamp errors." I can't vouch for the Swiss and Italians, but I would bet the keys to the kingdom that there is a computer glitch at the NYSE. High-frequency trading (HFT) is distorting the markets. It is enriching a few pockets (and that of the exchange), and I simply do not see how it is in the interest of the public to allow it.

I also know that fighting HFT is spitting into the wind, as faster tech comes along every few months. If you force the HFT funds to put their servers across the street (losing the time advantage of not being co-located with the exchange servers - milliseconds count!), it will only be a few years until technology has given the edge back to them. In ten years, when artificial intelligence and connection speeds are far more advanced, how will human traders compete?

Hire yet another AI to fight back? Wire yourself into the system (already being done, by the way, in rudimentary ways)?

The only way to effectively end HFT is for the exchanges to stop giving incentives for such trading. I can see the profits for the traders and the exchanges. I just don't see the benefit to the rest of us. The SEC should step in and settle some hash over missed time stamps. If a small broker-dealer has a wrong time stamp, they are all over us, and you can bet there are fines.

Something is wrong here. If one trade can go "back to the future" then how many more? Really?

E' necessario fare qualcosa per garantire che i mercati siano un ecosistema sano nel quale agenti con prospettive e scale temporali di investimento diverse possono coesistere senza distorsioni ed eccessive "asimmetrie tecnologiche". La Tobin Tax proposta da Merkel e Sarkozy non è però il modo corretto di affrontare i problemi: questa almeno è l'opinione di Jason Zweig nel video qui sotto (e anche in questo articolo sul WSJ di qualche settimana fa)

Gli algoritmi come forza evolutiva?

Profili di montagne che imitano indici di borsa: ecco un video sull'algorithmic trading davvero molto molto ben fatto tratto da www.ted.com.  Kevin Slavin ci invita a ripensare al ruolo della matematica nel mondo contemporaneo:

...I want to propose today that we rethink a little bit about the role of contemporary math -- not just financial math, but math in general.That its transition from being something that we extract and derive from the world to something that actually starts to shape it -- the world around us and the world inside us. And it's specifically algorithms, which are basically the math that computers use to decide stuff. They acquire the sensibility of truth, because they repeat over and over again. And they ossify and calcify, and they become real.

Oltre a ricordare il flash crash e le stranezze del trading ad alta frequenza documentate da
nanex nel video scopriamo che gli algoritmi che assegnano un prezzo ai libri venduti da Amazon
possono pure prendere delle robuste cantonate:

And the gag is that, of course, these aren't just running through the market. You can find these kinds of things wherever you look, once you learn how to look for them. You can find it here: this book about flies that you may have been looking at on Amazon. You may have noticed it when its price stared at 1.7 million dollars. It's out of print -- still ...(Laughter) If you had bought it at 1.7, it would have been a bargain. A few hours later, it had gone up to 23.6 million dollars, plus shipping and handling.And the question is: Nobody was buying or selling anything; what was happening? And you see this behavior on Amazon as surely as you see it on Wall Street. And when you see this kind of behavior, what you see is the evidence of algorithms in conflict, algorithms locked in loops with each other, without any human oversight,without any adult supervision to say, "Actually, 1.7 million is plenty."