In simple terms, an algorithm is a set of instructions which can be used to solve a particular problem. When used by computers, algorithms become incredibly powerful, allowing the machines to perform often complex calculations and tasks multiple times in a fraction of a second. Algorithms form the foundation of Artificial Intelligence (AI) and machine learning, whereby processors can be given a set of instructions and can then, via multiple high speed repetitions, ‘learn’ how to perform the task in hand more efficiently.
Although words and phrases such as ‘algorithms’, ‘machine learning’ and ‘artificial intelligence’ might make it sound as if any discussion of algorithms should take place in the lofty realms of top level computer programming, the fact of the matter is that algorithms play a massive part in the everyday life of virtually every person. From the price you pay for something ordered from Amazon to the box sets which Netflix recommends and the route suggested to you by Google Maps, algorithms are used time and time again to enable the smooth delivery of the kind of automated services which we’ve all come to take for granted.
The same kind of change has taken place in the realm of forex trading. Think back just three decades and the world of forex trading was radically different. Most trades were carried out over the phone, information on the value of currencies had to be painstakingly sought out and there was a huge disparity between the amount of knowledge that large institutions were able to access when making trades, and that which the smaller trader could utilise. In fact, back then most of the speculation involving currency was carried out by large institutions and well-funded speculators, with retail traders struggling to gain a foothold.
The arrival of the internet changed the course of the history of forex trading, creating an open, global market place which enabled trading to take place 24 hours a day, five days a week and ultimately resulted in a market which handles more than $5 trillion per day and dwarfs every other kind of market. The development and application of algorithms played a huge part in creating that market and, as the use of big data grows ever more pervasive, algorithms and the forex markets are set to become even more closely entwined in the future.
Trading using algorithms involves setting out a range of parameters on the platform being used, covering aspects of trading strategy such as the price of trades, timing and the amount to be traded, and then trusting the algorithms to carry out trades which reflect the chosen strategy. As well as simplifying the trading process, particularly for those who are new to forex, algorithms can be used in ways which allow greater speed and keep costs lower. Among the basic types of algorithm used by traders are the following:
an algorithm which analyses large amounts of historical data and uses this analysis to seek out trading opportunities based on the results of this analysis. Although past performance is not always a guarantee of future behaviour, the sheer scale of data which algorithms are able to analyse means that the trades they make are far more likely to adhere to the reality of the market, spotting patterns which are likely to elude a person attempting the same research.
using algorithms to limit your exposure to risk. When leverage is taken into account, any sudden shifts in the forex market can have a huge impact on the amount of risk a trader is exposed to. Algorithms can be used to set stop loss orders which close positions down as soon as a specific amount of capital has been lost, or to sell when profits hit a chosen level. Algorithms such as these protect individual traders and reduce the degree to which they have to monitor the positions they have opened. They also protect traders from sudden shifts in the market caused by the ‘herd instinct’ of traders around the world noting a position and shifting en masse.
in the past a trader would have to search the markets – often via a third party – to find the price of a pair of currencies before deciding whether to make a trade. The advent of algorithms means that any retail trader has access to a constant stream of prices which is updated in real time to reflect any changes in market conditions.
High frequency trading
the use of algorithms makes it possible for retail traders to make trades in milliseconds – speaking literally – taking advantage of tiny shifts in the price of currencies before the wider market spots this shift and adjusts to correct the position. These can also be combined with risk-transfer algorithms, which lock in an order and protect the trader from the risk of the market moving while the order is being placed. In most cases, this risk will be carried, instead, by the broker, which will often be a bank seeking to profit from the spread between the price at which the currency is bought and sold.
the ultra-high speed collection and analysis of data which is made possible by algorithms opens up the possibilities of retail traders taking advantage of arbitrage, which is the process of setting up positions to cash in on the slight changes in the value of currencies as quoted by different markets. In particular, algorithms make it easier to take advantage of triangular arbitrage, a technique which involves trading a currency through two different currencies and then back to itself, taking advantage of differences in the value quoted. Arbitrage of this kind is only really possible thanks to the speed of trading enabled by algorithms.
The future is set to see the algorithms themselves taking over many of the functions of the trader. It is already possible for traders to set specific instructions for algorithms to purchase a certain amount over a set timescale from specific liquidity providers, and the most advanced algorithms can be set to seek out trades during a specific time period, using market analysis to spot opportunities and machine learning to know who to purchase from.
The risk of the widespread use of algorithms is that it will reintroduce the traditional gap between retail traders and larger institutions, as the larger institutions will be able to invest heavily in the latest technology and execute trades at a speed which retail traders find hard to match. The other risk is that removing the human factor from trading will impact on the way the markets behave. While algorithms may be free from the tendency to respond in emotional terms to changes in market conditions, they may also be inflexible during times of high turbulence, carrying out trading strategies which cause the markets to shift even more dramatically. The future involves accepting that algorithms are here to stay and shaping their use in a way which maintains forex markets which are highly liquid, with easy access and low volatility.