The explosion of data, which has led to the birth of several new technologies, has revolutionized the way of functioning of every industry, and digital disruption is the new norm in today’s world. The dawn of digital is something that is being been realized by the financial markets as well currently. As far as finance is concerned, thanks to the extensive data available, trading is becoming even more technology-driven, data-driven and quantitative. This trend has led to making the markets more efficient, and has also created new jobs and roles for engineers. Along the same lines, Mr Prodipta Ghosh, VP, QuantInsti in an interview with DataQuest talks about algorithmic trading and the role played by engineers and technologist in the same.
Role being played by technologists and engineers in financial markets
Data and speed have always been integral parts of trading and investing. In the early history of trading, traders used fast boats, stagecoaches or even racing pigeons to gain an information advantage. But computer-based trading and technological innovations have steadily made speed an important and consistent source of alpha. The second phenomenon is an unprecedented explosion of data we are witnessing right now. It is believed that by 2025, the total volume of data existing globally will reach 175 zettabytes. This is forcing investors of all feathers to adapt their styles.
Trading is becoming even more technology-driven, data-driven and quantitative. On one hand, this involves adaptation of technology throughout the trade lifecycle. On the other, the ability to analyze a vast amount of (often unstructured) data brings forth promises of untapped alpha. This is where technologists and engineers bring indispensable skills to the table.
Engineers and technologists spend their formative years honing analytical and problem- solving abilities. These are highly transferrable skills for financial markets, apart from the relevant technological know-how. Of course, on their own, they are by no means sufficient for success. But with the current trend, these skills are fast becoming absolutely necessary.
Trading skills, Tech skills and Finance
Trading skills are a mix of soft and hard skills. For anyone from a tech background trying to invest for the first time, the biggest obstacle is information overload. The key is to develop a rough mental model of how the market works. This partly is achieved from knowledge acquired by a short and systematic study of market theories, economics and valuations.
There are many free and paid resources both online and offline – ranging from short courses to comprehensive modules on quantitative and algorithmic trading – which can help one get started. The second piece is behaviour development – ability to stay focused, respect discipline and assimilates concepts of risks. This is hard to teach and must be learnt. The best way to get started here is to get involved. Managing a small portfolio and following the market with some skin in the game can be a great help. The final piece is what is known as investing acumen – the ability to connect the dots, anticipating investors’ behaviour, and seeing the big picture. This is the hardest part and comes with experience, observation and time.
With the right mix of market knowledge, analytical abilities and focus, it is possible for anyone to excel in investing, irrespective of professional or academic background. The quantitative lean of investing these days, in fact, may even pose an advantage for beginners from a technical background.
The story on the other side is also improving. Today, it is getting easier for people without a technical background to pick up programming and data analysis as well. Python is a great starting point for a non-programmer. And with many tools and platforms that allow users to focus on goals rather than technical complexities, systematic trading is slowly being democratized.
Growing tech roles within Algo and Quant Trading
As I mentioned that the role of technology in investing & trading is increasing rapidly, it is even truer for Algo and Quant Trading, where technology is one of the key pillars along with knowledge of statistics & financial markets acumen. There are various profiles that Algo & Quant Trading firms have been actively hiring including:
- Infrastructure Developer: They are responsible to create and maintain the trading platforms, work on exchange APIs, create back-testing platforms, etc.
- Strategy Developer: Those who have a good understanding of the markets and can code their own or traders’ strategies.
- Network Infrastructure: They focus on ensuring high network uptime and acquire the lowest network latency possible to give the trading team the speed edge.
Given the importance of technology in Algo and Quant Trading, techies remain in high demand (and hence highly paid!) at all times.
How Algo trading has made markets efficient
Algorithmic trading has a mixed reputation among the public and there have been alleged cases of misuse in the past in many markets. But, in general, the conclusion on the impact of market efficiency has been positive. Academic reports suggest algorithmic trading has a causal effect on improved (lower) liquidity costs, reduced buy-sell imbalance and volatility. A large part of algo trading involves high frequency (HFT) market making where the algos are liquidity providers. This market is intensely competitive, and like all competition, it brings down the cost of goods (in this case, cost of liquidity) for the end consumers, i.e. liquidity takers (like most retail traders).
Many regulators voice concerns about the transience of this liquidity, but we have little conclusive evidence. Some researchers suggest even on the contrary, especially for the Indian market, that algo trading is associated with lower intraday liquidity risks and possibly lower extreme price movements (potentially driven by algo arbitrageurs). Depending on the market micro-structures, algo trading may have a mixed impact on certain types of liquidity takers. For example, large fund houses in fragmented venues (like the US market) allegedly faced higher trading costs. But in general, for retail investors across the board – who neither send large orders that have to be sliced and diced nor respond to micro-second market movements – algo trading has generally been beneficial. The steadily shrinking bid-ask spreads across the board in Indian equity markets is a case in point.
Tips on why people lose money to trading and how systematic trading is helpful
Trading is ultimately a game of uncertainty and even very smart people in large trading houses are not immune to trading losses. It becomes a problem only when we lose consistently, i.e. have a negative expected return from our trading. This, usually, is a result of incorrect behaviour and expectation. First, we need to set our expectations right. Trading is long term game, and a get-rich-quick approach will usually fail (unless one gets really lucky). Second, the style of trading has to be chosen carefully. A highly concentrated portfolio may be suitable for private equity investors, but can be ruinous for retails. Similarly, a retail trader will not be able to successfully execute latency sensitive quant strategies (like high-frequency arbitrage) competing with trading powerhouses.
The key is to choose the style well – like systematic risk allocation (which encompasses systematic value investing) or other suitable low or medium frequency strategies. Finally, it is important to be consistent and focused on generating trading signals and managing risks. Developing a systematic way to trade can be of great benefit here. Properly done, it can ensure that the idea generation process is based on sound hypothesis, thoroughly tested, and executed with precise discipline.
But systematic trading comes with its own caveats. The expectation, again, has to be set right. Systematic trading is no free lunch, nor is it fire-and-forget. It involves rigour in research and regular monitoring and analysis. Also, If not properly done, they are susceptible to over-optimized strategies that look wonderful in back-testing and blow up in real life. And finally, discipline that a machine brings is only as strong as the man controlling it.