Computers have taken over many facets of our lives which includes activities, which we wouldn’t have let them down a few decades back. One of those activities is letting them trade stocks, bonds and other financial instruments on our behalf. Now when I speak on computer-trading or more accurately algorithmic or automated trading, I am not referring to online trading, which I’m sure quite a few readers of this site is familiar with. Unlike online trading, algorithmic trading are carried out by sophisticated players such as investment banks, hedge funds, pension funds and other institutional investors.
What basically happens in an algo trade is that computer programs are used for entering trading orders with the computer algorithm deciding on aspects of the order such as the timing, price, or quantity of the order, or in many cases initiating the order without human intervention. A special type of such trading is High Frequency Trading, in which very powerful computers (usually supercomputers) make elaborate decisions to initiate orders based on information that is received electronically, before human traders are capable of processing the information they observe.
The magnitude of this type of stock trading is tremendous, both financially and ethically. It is estimated that nearly 70% of all US equity trades are algorithmic in Nature, i.e without a live person actually feeding in the orders. The tremendous increase in computing power means that these algos can can sift through an enormous amount of market data both historic and real time. They can then gauge trends and make buy/sell decision on certain parameters within the blinking of an eye and parse trades in millionths of seconds. These algos incorporate a fair amount of AI so that it can learn from past experience. Since most trading is algorithmic anyhow, some of these programs are cunning enough to detect other algos embarking on predictable trading strategies, and ruthlessly adjust their techniques. They’re growing ever more complex, subtle, and sophisticated. An example of this is Citibank’s “Dagger” a sophisticated algo trained in secret by Citi’s financial and investment experts, is capable of stalking through 20 different markets hunting for anomalies, buying and selling, prowling through mountains of historical data—all at the behest of Citi’s clients. Deutsche Bank also has a similar algo called Stealth and Sniper and Credit Suisse has Guirella. These algos work on arbitrage, statistical arbitrage, trend following, and mean reversion, all complex financial techniques to determine pricing and quantity of trade.
Stock exchanges are also getting into the act, and providing facilities such as low latency proximity hosting and colocation. Since latency can influence the the quantum and timing of information these algos can process, sophisticated investing groups spend a lot of money on the fastest data access from their computers to the exchanges, and that can make a significant difference in the way they make money from these trades. Even though algorithmic trading, bound by certain regulations, are legal, there are certain other types of trading which falls in the grey area of market manipulation. Flash Trading is controversial practice of some financial exchanges whereby certain customers are allowed to see incoming orders to buy or sell securities very slightly earlier than the general market participants, typically 30 milliseconds, in exchange for a fee. Now a 30 milisecond advantage is useless to a humar trader, but not so for ultrafast supercomputers which can execute millions of statistical analaysis and carry out high frequency trading in those precious milliseconds. This is the legal equivalent of “frontrunning”, an illegal activity which allows some traders to access and process information before the market, thus profiting from them.
It is also alarming to see how humans traders are being increasingly taken out of the equation. In a hedge fund startup called Rebellion Research, in New York, which was started in 2007 with $7 Million, a bunch of twentysomething math and computer whizzes designed and maintain a learning algorithm called Star. “It’s pretty clear that human beings aren’t improving,” said Spencer Greenberg, 27 years old and the brains behind Rebellion’s AI system. “But computers and algorithms are only getting faster and more robust.
What makes Star intelligent, says Mr. Greenberg, is its ability to adjust its strategy based on shifting dynamics in the market and broader economy. The program isn’t wed to any single investing approach. Under certain conditions, the fund will buy cheap stocks, in others it will favor stocks with swiftly rising prices—or both at the same time.The program monitors about 30 factors that can affect a stock’s performance, such as price-to-earnings ratios or interest rates.
The program regularly crunches decaded of historical market data and the latest market action to size up whether to buy or sell a stock. When certain strategies stop working, the program automatically incorporates that information, “learns,” and adjusts the portfolio. Every morning, Star recommends a list of stocks to buy or sell—often it offers no changes at all. A human trader implements the moves. The firm says it never overrules the computer program, which is largely the same system they started with in 2007. Star was designed with the simple goal of beating the broader market, something which it has consistently achiever since the firm was founded. Rebellion gained 17% in 2007, compared with the 6.4% gain by the Dow Jones Industrial Average. It stayed defensive throughout most of 2008, holding gold, oil and utility stocks. Still, it lost money like most investors, sliding 26% but topping the 34% decline by the Dow industrials.
In early 2009, Star started to buy beaten-down stocks such as banks and insurers, which would benefit from a recovery. “He just loaded up on value stocks,” said Mr. Fleiss, one of the founders referring to the AI program. The fund gained 41% in 2009, more than doubling the Dow’s 19% gain. The defensive move at first worried Mr. Fleiss, who had grown bullish. But it has proven a smart move so far. “I’ve learned not to question the AI,” he said.
Now just computers are efficient does not mean they’re infallible. Owing to the complex nature of financial markets, where the laws of predictability does not apply, the algos can often mess up and often spectacularly. For instance, on May 6, 2010 a huge crash occurred involving U.S. corporate stocks, followed by an almost immediate rebound. It was the second largest point swing, 1,010.14 points, and the biggest one-day point decline, 998.5 points, on an intraday basis in Dow Jones Industrial Average history. Such loss usually denotes the the start of a depression, but the massacre was over in 10 minutes, and rebounded back to normal levels. Here’s a chart :
Though no particular reason could ever be ascertained, but the most likely being automated trading errors. Despite the fact the automated trading have increased volumes and returns for savvy investors, high frequency trading, without adequate regulations and using leverage or boorwed capital can also can erode wealth and often can be counterproductive to general economic activity. At the end of the day, computers are merely tools, it’s human ingenuity and greed which drives them. Indian exchanges like BSE are also just beginning to enable automated trading, it’s high time that they should stop blindly aping the west and think about pros and cons before they implement, due to market pressures. This activity does not help the small retail investor and they are the people who have propped up the Indian economic miracle for the last 3 decades.
