In today’s fast-changing landscape, mathematical analysis is being used to explain the world around us, and quantitative techniques are becoming ever more widespread. The rise of quantitative trading means the financial markets are no longer dominated by fast-talking Wall Street brokers. Instead, the world of finance is now a highly diverse environment where top mathematics, statistics, computer science, engineering and economics graduates all work to apply their skills and make things happen.
Quantitative trading involves using quantitative methods and algorithms to execute strategies. This has a broad set of uses; a typical example might be a trader using a mathematical model to take a position on what an asset “should” be worth before carrying out a trade. Models like these are designed to leverage information the market may be missing, such as what happens to a company or an industry when interest rates move in a particular direction. However, this is only one form of quantitative trading, and is most often performed by hedge funds or investment banks.
At IMC, quantitative trading is used for what’s known as “market making” – the act of providing liquidity to the market. Rather than leveraging information asymmetry, the strategies revolve around making it easier for participants to buy and sell assets. This keeps the markets running smoothly, especially when things are volatile or if the market is small in scale.
As a market maker, IMC uses quantitative strategies to continually provide two-sided quotes. This means that, rather than deciding whether an asset is over- or under-priced, IMC instead provides a price to buy and to sell for each given asset. To use a simplified example, say you visit your local corner shop and you find a bag of peanuts for EUR 3.50 – the shop’s providing a one-sided price quote, namely the sell price (or the “ask” price). But, instead of buying, what if you were looking to sell a bag of peanuts? Ideally, your local store would also be buying peanuts for a listed price of, say, EUR 3.00 a bag – a.k.a. the “bid” price. At this point, the store is no longer acting as a traditional shop; it’s creating a marketplace where people can both buy and sell.
Similarly, IMC’s role is to take both a buying and selling position on specific stocks or other assets. While a trader at a managed fund may have the discretion to take a more active position (they might use a model to determine a specific asset is worth a certain amount), IMC doesn’t take any set positions itself, but instead uses quantitative trading to minimise or manage risk.
For market makers, margins are derived from carrying out multiple buy and sell orders each day, minute after minute – and eventually making the spread between the bid and ask prices when things go well. However, to mitigate the risk involved, IMC must create – and continually fine-tune – a range of quantitative models.
IMC generally uses three types of algorithm:
To make things easier for market participants, IMC discovers and aggregates the positions of buyers and sellers on several different exchanges. Valuation algorithms are designed to figure out the price of an asset based on the information provided by these exchanges, and then convert this into the right buy and sell quotes.
Although IMC doesn’t take a set position on an asset’s value, it maintains a running inventory of stocks and options in order to always facilitate smooth trading. Holding these positions carries significant risk because subsequent price changes can easily leave a market maker out of pocket.
Another risk in providing liquidity is being open to “adverse selection” – which is the possibility a buyer or seller knows more about an asset than the counterparty (in this case, IMC). This means other participants might be better informed about an asset and use that information to win trades. Position management algorithms are developed and used to reduce or manage this risk.
Execution algorithms carry out trades and manage orders while taking market developments into account at all times. Many options, for example, depend heavily on prevailing stock prices – so when the stock price moves quickly, the option prices move with it. If there’s a big move and a trader isn’t fast enough to cancel an order before someone else executes, the market maker could lose money. Knowing when to cancel a buy or sell order is just as important as knowing when to carry it out –execution algorithms are used to make these decisions in an instant.
While quantitative traders have a more operational role, quantitative researchers build and maintain the models and algorithms used in trading. A trader uses the algorithms to execute a strategy and has the discretion to change the parameters as they see fit. However, if a strategy isn’t getting the desired results for some reason, a researcher can step in to see what’s going wrong. He or she can educate the trader or improve the algorithm – either by changing its behaviour or by augmenting it.
At a hedge fund, a quantitative researcher tends to spend nearly all their time sifting through a mountain of data to try and find “signal” – which is a trigger to buy or sell on an advantageous trade. At a market maker like IMC, on the other hand, the researcher’s job to help reduce the “adverse selection” the market maker is exposed to – with the added challenge of having to also maintain continuous two-sided price quotes. In this sense, a researcher at a hedge fund is often operating on the other side of the coin to a researcher at a market maker.
Quantitative trading isn’t just for traditional Wall Street types. In fact, it’s about as close as it gets to having a full-time – and well-paid – career in applied maths. This makes it well suited to ambitious graduates who love working with mathematics, statistical analysis, computer science or economics, and enjoy spending their days problem-solving and overcoming complex challenges. If you’re competitive and passionate about science and maths, you’re likely to be well suited to a quantitative role. It also helps to be adaptable, because algorithms and strategies need to be fine-tuned and developed continuously. Even if the model is perfect, it still has to be adapted every time the market changes.
Think you’ve got what it takes to work in high-frequency trading? Learn about the opportunities on offer at IMC via our graduate careers site.