Jobin Reji, King’s College London
So, we have looked at how Artificial Intelligence is being implemented in fintech on a general scale. In this article, and onwards, we are to look at how AI is being used in specific sub-sectors starting with Algorithmic Trading. What is Algorithmic Trading you may ask; according to Investopedia, ‘Algorithmic trading (also called automated trading, black-box trading, or Algo-trading) uses a computer program that follows a defined set of instructions (an algorithm) to place a trade.’ Note, there is a difference between automated trading and Algo-trading whereby automated is just the executions humans were already carrying out whereas Algo-trading involves creating orders based on models and strategies that are constantly being improved by the algorithm.
Algorithmic trading AI eradicates much of the human error by generating profit at a speed and frequency that humans are not capable of. What does this mean? Well due to advance capabilities, AI has the power to execute trade orders in precise timings that profit is nearly always guaranteed, it takes much of the risk away as with trading, even the millisecond can significantly hinder returns. However, this does make the market more volatile as it rules out the impact humans have so you can expect to see large dips and steep rises if the trading floor of every bank were to be algorithmically traded.
Algorithm’s Real Life use
Algorithmic Trading is widely used in Investment Banks, Pension Funds, Mutual Funds and Hedge Funds. Algo-trading is particularly used in foreign exchange (Forex) trading as this market is seen as proportionately more volatile than any other recognised market hence the use of AI is most efficient. What this means for traders is that they no longer need to monitor live prices or graphs, nor do they have the need to put in orders manually.
Could this possibly mean redundancy for many traders? Well, in the United States, one of the largest markets with the largest economy in the world, 80% of their trades, in 2019, was executed using machines. However, algorithmic trading is mainly being targeted at short-term markets. The use of Algo-Trading is also used widely in the energy market with the ever-changing views on non-renewables to zero-carbon start-ups, the energy market is being revolutionised as we speak. Due to this, human error in energy trading has increased in this specific market with psychological influence playing the largest role in many faults. No matter how much experience a trader has, they can’t eradicate any biases or preconceptions, unlike the algorithm. Nevertheless, it won’t ever be perfect with risk ever-present. Network problems, bugs, imperfect algorithms, and data reading errors can have large effects on trading outcomes; problems that aren’t viable through humans thus we won’t see any replacements any time soon.
Algorithmic trading is not here to replace traders but to work alongside and improve their judgements skills and financial knowledge. This allows the trader to become more efficient in their work as it allows them to adapt to volatile markets and see real-time effects of trade executions. The more complex an algorithm, the more stringent backtesting is needed before it is put into action.