Hands-On AI Trading with Python, QuantConnect and AWS :: InvestMacro

AI is all the rage these days. We know this! But as investors and traders, do we know how to incorporate AI into our systems? Do we even know the many possible ways we could use AI to help our trading? Well, today I am going to do bring something a little bit different to the blog, a quick book review!

As a Python coder, automated trader and investor, I feel constantly bombarded with bits and pieces of AI trading information from newsletters or ‘how to’ tutorials to implement this or that. Luckily, I was recently given a complimentary copy of Hands-On AI Trading with Python, QuantConnect and AWS and, it turns out, this book is a comprehensive guide that brings a whole lot of information into one place with a consistent presentation and coding style.

Basic Information:

This book was written by five active data-driven market professionals that all run businesses or have positions that are aligned to the financial markets and/or using AI and automated solutions. Jiri Pik is the CEO of RocketEdge.com, Jared Broad is the founder and CEO of QuantConnect, Ernest Chan is the founder of PredictNow.AI, Philip Sun is the CEO of Adaptive Investment Solutions and Vivek Singh previously worked at a hedge fund and is now a senior product manager at AWS.

This book is targeted towards those in finance, aspiring quants, veteran quants, hedge fund traders, as well as independent traders & investors. As you can tell from the book’s title, there’s a focus on using the Python programming language as well as the services of QuantConnect, Amazon Web Services (AWS), and Predictnow.ai.

The authors present these specific tools (QuantConnect, AWS, Predictnow.ai) as a tech-stack to get things from start to finish. As stated in the book, the goal was to provide, “an easy-to-setup and use environment where readers could instantly experiment with the algorithms to build their confidence without spending any time setting up the required infrastructure.” In other words, the reader has an opportunity to go from the learning, creating and testing phase (with code and AI models) to potentially working through to a live strategy trading (through QuantConnect and their connected brokers).

I found the book to be well organized and it is structured into 3 main parts.

Part 1 is about the Capital Markets and Quantitative Trading.

Part one quickly brings those unfamiliar with the financial markets up to speed. It covers various topics from the different types of markets traded to the mechanics of how things work in the market ecosystem. This includes all the different types of participants, the different roles they play, the different types of orders these traders use as well as who has unique types of informed access. The authors go further through derivatives, futures, charting, crypto and more.

The quantitative analysis and trading part of this section brings a comprehensive overview of quantitative trader functions using QuantConnect and Python code. It details the steps, processes, and aspects that quants will go through, experience and need to consider for a successful process. I think this section will be very beneficial for aspiring and seasoned quant traders alike, as this book does a great job of laying out the market framework and the quantitative trading landscape.

Image from example in Hands-On AI Trading.

Part 2 goes into AI and Machine Learning (ML) in Algorithmic Trading.

Part two focuses on AI-based algorithmic trading. Here, you start to address the market prediction, forecasting or other specific problems you’re trying to solve. You proceed step by step, breaking down issues and finding solutions using AI and machine learning processes. It details the data set preparation, handling data, creating features, and splitting datasets into training and testing phases.

If you are unfamiliar with AI models – this section (especially Chapter 4) is for you as it delves into models like linear regression, Markov, Bayes, decision trees, support vector machines, neural networks, and many more. Found alongside these characteristics and concepts is the Python code you can use for these different types of quant functions.

Part 3 delves into Advanced Applications of AI in Trading and Risk Management.

Finally, part three discusses using these AI models in real trading and investing scenarios. The authors provide 19 specific examples and this is where I think the main strength of this book lies. These examples illustrate different aspects of the investment game or problems that are solved using various AI models for major financial markets (FX, stocks, etc.). These examples, once understood, ideally can form the basis for many new ideas, as well as just understanding how these pros go about it. Also, the Python code is included for these examples.

For instance, one of my favorite examples (#8) was just a simple exercise in using a stop-loss based on historical volatility (and drawdown recovery). This example used a LASSO regression model with features including the VIX, Average True Range (of n months) and Standard Deviation (of n months). The example used a few different methods to test variations of a dynamic stop-loss order to varying degrees of success. This type of example represents a common problem most traders come into when working through their strategies.

The examples also give interesting ideas on how to use AI and models in use cases beyond just trying to predict future price returns.

Overall Takeaway: 

I thought this book was well done and is the best book that bridges quant trading and AI together that I have read so far. I think a lot of the AI and machine learning aspects were explained and guided in a clear, concise, and a well-organized way, since it’s very easy to get lost in the weeds with this subject.

The breadth of coverage among these many strategies, concepts, and factors involved is admirable, covering all the way from data acquisition and programming to the role of generative AI. There’s a lot to unpack. There’s a lot to learn. I think it’s a testament to the authors that they created a book that covers so much. There’s also a github repository for the examples.

I would recommend this book for any aspiring quant traders or programmers, or anyone who is interested in the understanding of these markets, especially in how quant trading and AI intersect. I would also recommend it for traders looking for examples of AI in trading or finding new ideas to implement AI strategies.

Disclaimer: Complimentary book copy was provided by Wiley.

Article written by Zac@InvestMacro