Electronic & Algo Trading: Make or Buy?
Automated trading refers to the use of computer technology and the Internet to trade in electronic markets. It transformed trading in the main markets for equities, fixed income, currencies, commodities, and derivatives.
Nowadays, hedge funds and other traders can auto-trade financial instruments on public exchanges such as NYSE, LSE, CME and LIFFE, and other venues offering electronic trading, via third-party platforms or direct market access (DMA).
The simplest form of automation comes from using stop and limit orders, and opening or closing positions when a set price level gets hit. The most complex form employs quasi-AI based (such as machine learning).
Technology infrastructure
Older electronic trading venues (the public exchanges above) tend to develop their own systems, while some newer venues are happy to use third-party software. They offer two access options:
- GUI – traders run an exchange application on their desktop to connect to the venue
- API – traders plug their own platforms directly into the venue
Traders can analyse markets in-depth and place orders using third-party platforms such as Bloomberg or Reuters, or their own platforms.
The Financial Information eXchange (FIX) Protocol supports the process of connecting counterparties. It is the standard financial organisations use to communicate trade information.
Algorithmic trading
More and more electronic trades are generated by hedge funds and other trading firms running fully automated systems. Which let machines manage all trading processes, including what to trade, when and where. These systems employ algorithms. Some of them are relatively simple, some are quite complex.
In trading, an edge is a method that enables to accumulate more profits than losses over time. In order to gain an edge, many hedge funds are employing “super algos”, quasi-AI based trading strategies. Some of these algorithms search for patterns, adjusting to what works in the markets that day, week or year.
Algorithms are used in many trading strategies. A lot of algo-trading today is high-frequency trading (HFT), which takes advantage of short-term opportunities, leveraging automation to buy and sell financial instruments in short time spans.
Make or buy
Developing an automated trading platform in-house is expensive to develop and manage, especially if the trading strategies require access to several exchanges, complex algorithms and people with advanced degrees in subjects such as Mathematics, Physics and Statistics.
There are third-party solutions which are cost-effective and tried & tested. The make or buy decision depends, among other things, on trading strategies. Strategies include Momentum, Statistical Arbitrage (Statarb) and many more.
Final thoughts
In the late 2000s, I was working for a UBS’ algo trading desk in London, dealing with the above and more. Since then automated trading got bigger (although the fun seems to be mostly outside the large investment banks now). It got more complex too. However, whether you make or buy the trading platform, one thing has not changed: unless you have an edge, you will never make money.
To learn how we can help hedge funds and other assets managers considering a move into automated trading:
