Fund managers do their maths
More than 28 per cent of all US equity trading was driven by so called "algorithmic trading" at the end of last year, as traditional fund managers followed the lead of hedge funds and began to embrace complex automated trading strategies.
The figure is set to rise much further in the coming months and years as fund managers (along with brokers and exchanges) strive for ever-greater efficiency and control over the trading cycle amid heightened market competition and consolidation.
Algorithmic trading uses mathematical models, which are being developed by quantitative PhD-armed staff at a growing number of investment banks and at specialist trading firms, as a means of trading large blocks of shares quickly.
Algorithmic traders decide when to place buy or sell trade orders on the basis of quantitative models that automatically generate the timing and size of orders. These are based on goals specified by the parameters and constraints of a mathematical algorithm - a step- by-step mathematical problem-solving procedure.
Greater use of algorithmic trading does not come cheap. Last year, approximately $230m was spent across the US equity markets on the various technological components that make up algorithmic trading, according to NeoNet, the agency broker. By 2008, NeoNet forecasts the figure will reach more than $300m.
Programme trading - the trading of large baskets of stocks in an automated process, which is widely seen as a decent proxy for measuring the growth of algorithmic trading - represented nearly 60 per cent of total trade volume in 2005 compared with barely more than 10 per cent in 1989.
Perhaps the most common algorithm used by traders is the VWAP (volume-weighted average price) algorithm. This reflects the total value of transactions in a stock divided by the number of shares traded in a particular session. It has been used by pension funds for many years to evaluate performance.
Beyond the most simple algorithms, procedures can be developed for all manner of trades and can include an almost endless range of parameters designed to include a myriad market conditions and scenarios in the order generation process.
As the accent remains on automation, some fund managers worry about how algorithmic trading providers can differentiate themselves if all have access, provided they are able to pay for it, to the same technology.
One trader said: "Even a straightforward VWAP algorithm can deliver outperformance if the predictive curve is based on an optimised historical profile, or if it employs advanced queue management techniques. The most advanced strategies designed to minimise impact can demand huge computing power in order to measure correlationswith thousands of other prices.
"In a crowded market where trading needs change by the minute, there is a constant need to enhance and refine the way algorithms work.
"Market, sector and stock liquidity, and volatility are changing all the time, and the algorithms have to be constantly tuned to reflect this."
Richard Johnson, of Miletus Trading, a quantitative broker-dealer, said: "You need to be constantly tweaking algorithms to suit market conditions or a client's particular style of trading."
He added that merger activity in the exchange space was likely to speed the take-up of algorithmic trading still further as the New York Stock Exchange's proposed deal with Euronext and Nasdaq's recent stake-building in the London Stock Exchange appeared to be predicated on greater use of electronic trading.
Nasdaq recently launched a market information service, called Market Analytix, that provides an early insight into market moves before they actually translate into stock price or volumes swings.
The service is particularly geared toward algorithmic traders and buy-side participants who base their strategies on software programs that detect changes in the marketplace. This can then generate buy and sell orders.
Driving the heightened use and demand are asset management firms, particularly hedge funds, seeking more cost effective and efficient ways to trade, and sell-side firms looking for cost reduction.
Analysts attribute the greater use of algorithms by hedge funds to the fact that they tend to be more quantitative in their execution strategies and have fewer resources, so making greater use of automation. They also have a greater focus on costs and are more concerned with anonymity.
Tabb Group, the US consultant, said the next generation of equity-oriented trading algorithms would be more flexible, react to unexpected events, handle complex relationships, and begin to manage part of the investment process. |
