What is Quantitative Trading?
The first attempts at creating what would now be called a stock market occurred in 17th century Amsterdam with the goal of financing European colonial efforts, and ever since buying and selling shares of stock on an open market has been a cornerstone of Western society. In stark contrast to most of the history of trading though, where trading was limited to the very wealthy, the digitalization of the market over the past decades now means that virtually anyone with internet access and two dollars to rub together can participate in the stock market. This technological democratization has given rise to an unprecedented opportunity for individual traders to profit from the market.
The majority of stock trading throughout history has been based what we now call fundamental analysis. If you thought a company had good fundamentals - sound business model, market opportunity, experienced leadership, etc. - you bought stock. If you thought the opposite, you sold stock. Fundamental trading (generally just called "investing") is alive and well today - best characterized by Warren Buffet and your parent's IRA.
However, starting in the early 20th century and gaining tremendous traction since the 1980s is a different framework used to make trading decisions: technical analysis. This approach ignores the company fundamentals, and instead simply looks at a stock's "tape" - the price and volume of the stock's trades and how the current price of a stock changed over time ("price action"). Some very clever mathematicians figured out that they could calculate various statistics of a stock's price action (technical analysis) and build strategies that use those statistics to determine the optimal times to buy and sell stock - to great profit. And so a new type of trader was born: the quantitative trader. Unlike their predecessor, the work of the quantitative trader consists, not of analyzing a company's fundamentals, but of developing strategies that tell them what and when to trade. Much ink has been spilled debating the relative merits of fundamental vs quantitative trading, and we will explore that question in a future post.
These strategies are ostensibly quite simple: either buy or sell a certain amount of a certain stock at a certain time. If the strategy buys a stock at one time and sells it later at a higher price, the trade is a profitable one. The trader's goal when developing strategies is to maximize both the profit of each individual trade and the number of profitable trades executed. But the devil is in the details, and the unpredictable nature of the stock market can make formulating profitable strategies extremely difficult. Therefore specific technical analysis is insufficient to create quality strategies - the trader must have a robust and useful model of the market as a whole in order to use their technical analysis efficiently.
Another innovation that rode on the coattails of the digital revolution was the idea of trading automatically - why manually enter trades when you can program a computer to do it for you? This means that along with developing their strategies, the quantitative trader can encode these strategies as algorithms executed by a computer program. And so quantitative trading can be described as three concrete tasks:
- Building a overarching model of the market - how and why it behaves the way it does in order to discern underlying patterns in price action
- Building strategies that use this model to signal specific trades under specific conditions in order to maximize profit
- Building a software platform that implements these strategies and automatically executes trades
At its core, quantitative trading is about understanding and building - it requires both scientific and engineering efforts to succeed. The Meno Project focuses on the first two of these tasks, both because they are more difficult than the third, and because the building of these models and strategies is where for the real creativity inherent in quantitative trading lies.