Introducing algorithmic trading

There are various definitions of algorithmic trading, so we need to first clarify what we mean by this term in this book and what the perspective of these tools is from the viewpoint of professional users.

To achieve this, we are now going to cover the following:

• A definition of algorithmic trading
• What a quantitative hedge fund is from an organizational point of view

Algorithmic trading definition

According to a very common definition from Wikipedia, “Algorithmic trading involves using computer algorithms to help traders execute trading strategies based on factors like price, volume, and timing. The main advantages of algorithmic trading are speed and efficiency, backtesting capabilities, reduced emotional bias and diversification.”

Such a definition reflects what most people think about algorithmic trading but it doesn’t fit exactly the professional environment view. Let’s see why:

• First, let’s clarify execution. Using algorithms and automating trade execution are two different matters. We may use algorithms just to automate a research process to analyze the market or to develop tools to monitor the market. We can also use algorithms to fully auto-execute trades but, as you will learn, this is just a component of a wider trading project.
• Let’s clarify the concept of speed. The primary goal of algorithms is not always to make things faster; rather, it’s about providing support in the decision-making process and solving problems efficiently and effectively. In addition, fast trading is not suitable for individual traders, as it requires huge investments in technology and, therefore, it is a battlefield where big companies operate.

  In the coming chapters, we will learn how to choose a wiser approach for trade execution than increasing trading speed.

• The third misunderstanding is about emotions. At a certain point in the trading decision process, emotions will kick in. If you are a discretionary trader, emotions will kick in when you click the Buy/Sell button. If you are an algorithmic trader, they will kick in when you activate the ON/OFF system button. Emotions are just delayed, not removed. Algorithms will help you approach the market rationally and scientifically, increasing your confidence level, and thus reducing your stress during the trade, but you will never get rid of emotions.

In Figure 1.1, you can see how algorithmic trading tools can be classified:

Figure 1.1 - Algorithmic trading tools by trading speed and automation level

In Figure 1.1, we classify algorithmic trading tools by trading speed (ranging from a few ticks to monthly time frames) and automation level (ranging from 0 to 100%).

As you can see, there are several ways to deal with such tools. Apart from discretionary trading and high-speed tools, which are not suited for individual traders, this book will help you do the following:

• Program analysis tools such as statistical tools and machine learning frameworks
• Program monitoring tools, such as real-time trading dashboards and equity market scanners
• Program mechanical trading tools such as execution macros
• Program trading systems with a full automation approach

In conclusion, when we use the term algorithmic trading in this book, we mean any computer-based tool able to help traders analyze, validate, monitor, and execute trading strategies.

Algorithmic trading in quantitative hedge funds

When we talk about algorithms, we often forget that they are going to be used by humans. Therefore, I think it is essential to briefly describe my experience in the finance sector, where I worked for small quantitative hedge fund start-ups.

A quantitative hedge fund generally originates from the idea of a trader who, after several years of experience in the markets, decides to start a hedge fund business. Subsequently, they partner with other individuals who will handle the legal, administrative, and commercial management of the company.

In this initial phase of the hedge fund start-up, it is generally not a priority to build a proprietary trading platform. Hence, such funds find it very useful to utilize platforms such as TradeStation, which are ready for use. So, these funds generally temporarily hire a specialist familiar with EasyLanguage (TradeStation’s programming language) to transform the senior trader’s market insights into algorithmic tools that can help the entire team in the trading process. Experienced traders generally have a visual approach to the market, relying on charts as their primary tools, which they have been using for decades. Therefore, the EasyLanguage specialist is partnered with the traders for a certain period, seeks to align with their approach, and then constructs trading tools based on that understanding. Depending on the trading strategy, sometimes such funds hire a traders’ team: let’s imagine a trading room with 5–10 workstations, where traders work following the same dashboard that provides the team with the market perspective of the head trader. Simultaneously, they may have the freedom to act within well-defined risk rules. In other words, this type of organization leverages the computational power of algorithms combined with the individual sensitivity of the human trader.

Some other funds are a one-man band, and the trading decision process is very similar to the individual trader’s—except for the investors’ pressure and compliance matters.

At times, these funds create a small development department to continue research while the rest of the organization operates in the market.

In this way, TradeStation’s EasyLanguage becomes an indispensable tool both in research and development activities and in daily trading operations.