Embracing the future: How automated trading is reshaping the markets

October 18, 2016 | Rocco Savage

Traders are increasingly relying on high-speed data and intelligent algorithms to adapt to fast changing markets. This article takes a look at the growth in algorithmic trading and what it means for gaining an edge in markets moving forward.

London-based trader Nathan Rothschild was once the wealthiest man in Britain. He is believed to have sent carrier pigeons to the scene of the Battle of Waterloo. These pigeons brought him the news of Napoleon’s surprise defeat before anybody else knew it in London. This enabled him to buy a large amount of British bonds before the sudden hike in price as a result of the news that Britain had become the dominant nation in Europe. By the time the market had absorbed the news and the bond prices skyrocketed, he had already minted big pounds.

Timely access to market intelligence and swift trade execution has always been the key to wealth creation in capital markets. Today, traders rely not only on high-speed information networks for timely access to news and market data, but also on algorithms to swiftly generate and execute trading decisions.

Algorithmic (algo) and high-frequency trading (HFT) has attracted investors and regulators globally over the past several years. According to a recent report by market research company Technavio, the global algorithmic trading market is expected to grow 10.3 per cent CAGR between 2016 and 2020.  In India, algo-trading contributed 15.49% while co-location accounted for 22.85% (38.34% combined) of the total equity cash market trade volume of NSE in July 2016.

A common algo-trading strategy based on statistics is Trend Following. The idea is to follow trends for technical indicators such as moving average. A dual moving average crossover trading system uses two moving averages, one short and one long. A buy signal is generated when the short-term average crosses above the long-term average –which usually indicates the beginning of an uptrend.

Statistical Arbitrage is a strategy to leverage any temporary inconsistencies in the prices of securities. Often, the stock prices of companies in the same sector or similar businesses are closely correlated. A pair trader observes the relationship between two stocks and buys or sells whenever the relationship gets out of sync. The expectation is that the relationship would revert to its mean in the mid to long run.

Speed is of essence. The key is to execute the trades before other traders armed with similar information and algorithms can do so. Today, HFT executes trades in sub-milliseconds. Market participants aspire for solutions that offer low latency. Therefore, algorithms that can compute approximate values faster are preferred. Online one-pass algorithms can analyse large amounts of data every microsecond to generate useful trading signals.

Statistical models require constant tweaking, as markets continue to evolve. Hence, algo-traders are increasingly looking at artificial intelligence solutions to build more dynamic models that can adapt to change markets. In addition to market data, these smart systems can utilise alternative data such as news, weather updates, satellite images, videos and geo-sensor data to predict market trends. Moreover, equipped with deep learning, reinforcement learning and genetic algorithms, these systems ..

 

Join the Discussion