views

Algos boost your trading but always read the label

WHERE professional traders venture, retail traders won’t be far behind, anxious to pick up on the latest developments in technology and strategy. Algorithmic trading is no exception – there have been recent signs that individuals are more and more interested in using algorithms in their foreign exchange trading. Deutsche Bank’s retail foreign exchange trading arm dbFX went from minimal algorithmic trading volumes at the start of 2009 to 20 per cent by the end of the year. FXCM has just launched its programming services division, which codes algos for clients based on their trading ideas, in response to client demand.

So what is algorithmic trading and why has it piqued retail traders’ interest? Algorithmic trading is essentially a rules-based approach to trading the markets, which will use pre-defined rules built into the programme to decide the when and the how of placing a trade in the market such as its timing, size and price.

There are three main reasons why currency traders are choosing to trade using algorithms. Firstly, individuals who primarily had passive investments in other asset classes are now looking to take a little bit more control of their finances but don’t want to sit in front of a computer screen all day every day.

Secondly, improvements in technology are increasingly being rolled out to retail traders as well. Jeffrey Lins, executive director of quantitative research and algorithmic trading at Saxo Bank, says: “Application programming interfaces (APIs) are becoming more ubiquitous and more available at the retail end. The actual computational power of a PC is aiding the creation of programmatic environments that are available to the retail trader.”

Thirdly, the importance of high frequency data and high-speed transmission of trades and figures has meant that a fast response to data releases is required to trade today’s currency markets.

There are certainly benefits to leaving your trading in the hands of a computer programme. It removes the emotion from your trading, says Stewart Lancaster, head of operations at retail forex provider AIFX. By predetermining the risk parameters of your trade, you don’t get swept away by the markets and put on trades you can’t afford, he says. It also means you don’t have to be watching the screens intently – although checking in at least once a day to make sure everything is running relatively smoothly is recommended.

Traders who use algorithms may also be looking to supplement their ideas, says Betsy Waters, a director at dbFX. “They may run one portfolio with their own ideas and another one with an algo that they purchased.” But while enthusiastic retail traders might be keen to develop algos to work using more institutional-style technology, those buying algos from third parties should do their research.

Luke Quinn, a programming specialist at FXCM, says that one of the commonest mistakes is retail traders not using algos properly. “Algorithms are often specific to different types of markets and users need to recognise the type of market and then apply the right algo. Some algos are designed to trade all markets whereas some are more geared to one market rather than another,” he explains.

He adds that traders should test algorithms by using small increments in a live account because demo accounts and back testing are not 100 per cent indicative. “Demo accounts have perfect liquidity so you don’t have to worry about the depth of the market and back testing tools sacrifice some accuracy in the name of speed,” he adds.

Also, some algorithms work well at first because they are optimised for the prevailing market conditions. A few months down the line, however, the markets have moved on and the algo is no longer as successful. A piece of technology, as ever, is only as good as the person using it.