Much like with any other difficult endeavor, it is a good practice to read about the experience, recommendations, and explanations of people with extensive experience and a proven track record in the field. This is especially so when it comes to the financial industry and algorithmic trading in particular.

It would be a lost cause to try to discover each and every single insight, issue, and solution on our own. Due to this, what follows is a curated list of books that I consider to be exceptional when it comes to covering topics related to algorithmic trading.

BookAuthorOverall DifficultyMathematics RequiredProgrammingKey Topics
Advances in Financial Machine LearningMarco Lopez de PradoAdvancedAdvancedPythonMachine Learning, Fractional Differentiation, Cross-Validation, Feature Importance
Algorithmic TradingJeffrey BacidoreIntermediateBasicNoneMarket Microstructure, TWAP, VWAP, Optimal Order Execution
Machine Learning for FinanceJannes KlaasBasicIntermediatePythonNeural Network, Natural Language Processing, Time Series Analysis, Reinforcement Learning
Python for Algorithmic TradingYves HilpischBasicBasicPythonVectorized Backtesting, Event-Based Backtesting, Live Trading, Websockets, Capital Management
Algorithmic TradingErnie ChanIntermediateIntermediateMatlabBacktesting, Mean Reversion, Momentum Strategies, Risk Management
Systematic TradingRobert CarverBasicBasicNoneHuman Biases, Position Sizing, Trading Rules, Forecasting, Portfolios

Best Books for Algorithmic Trading

Advances in Financial Machine Learning – Marco Lopez de Prado

This book is packed with valuable information regarding the many mistakes that practitioners, even experienced ones, make when using machine learning for algorithmic trading purposes. Additionally, the author also presents academic research he himself published, but in a more intuitive way.

Much like the name of the book suggests, this book is by no means an introduction to using machine learning models in the financial industry. It won’t cover the basics of training a neural network with OHLC data. Instead, it is aimed at readers who had previous exposure to the subject in order to understand the merit behind the ideas proposed throughout the book. 

His techniques are not only very interesting but also controversial due to the fact that they are cutting edge and aim at replacing decades of research incorrectly done (according to him). Whether one agrees or not with his proposed techniques, the ideas are solid and do indeed deserve to be considered seriously.

Just to name a few, instead of creating price bars in a chronological way, he proposes aggregating them in a volumetric fashion. Instead of researching through backtesting, he recommends doing feature importance analysis. His ideas regarding fractional differentiation are of special interest for traders with at least a minimum understanding of time series models.

You can take a look at the following exposition of Marco Lopez de Prado himself in order to have a grasp of the key aspects of his thesis.

Also, in order to provide a quick overview of the topics covered, these are the chapters of the book:

  1. Financial Machine Learning as a Distinct Subject
  2. Financial Data Structures (Data Analysis)
  3. Labeling (Data Analysis)
  4. Sample Weights (Data Analysis)
  5. Fractionally Differentiated Features (Data Analysis)
  6. Ensemble Models (Modelling)
  7. Cross-Validation in Finance (Modelling)
  8. Feature Importance (Modelling)
  9. Hyper-Parameter Tuning with Cross-Validation (Modelling)
  10. Bet-Sizing (Backtesting)
  11. The Dangers of Backtesting (Backtesting)
  12. Backtesting Though Cross-Validation (Backtesting)
  13. Backtesting on Synthetic Data (Backtesting)
  14. Backtest Statistics (Backtesting)
  15. Understanding Strategy Risk (Backtesting)
  16. Machine Learning Asset Allocation (Backtesting)
  17. Structural Breaks (Useful Financial Features)
  18. Entropy Features (Useful Financial Features)
  19. Microstructural Features (Useful Financial Features)
  20. Multiprocessing and Vectorization (High-Performance Computing Recipes)
  21. Brute Force and Quantum Computers (High-Performance Computing Recipes)
  22. High-Performance Computational Intelligence and Forecasting Technologies (High-Performance Computing Recipes)

Algorithmic Trading: A Practitioner’s Guide – Jeffrey Bacidore

This book is by all definitions unique in regards to the contents it covers. Unlike other books with similar titles, Jeffrey Bacidore honors the traditional definition of “Algorithmic Trading” and teaches about the different ways hedge funds and large institutions execute orders under transaction cost constraints.

The book starts by laying out the basics of financial market microstructure that any serious algorithmic trader should be aware of. Later, Bacidore covers the most widely popular order execution algorithms that are currently used as benchmarks, such as the Time-Weighted Average Price (TWAP) and Volume-Weighted Average Price algorithms.

In later chapters, he covers more sophisticated execution algorithms and sequentially covers the measurement of its performance.

Jeffrey Bacidore is the founder of the Bacidore Group, a research and consulting firm specializing in trading-related topics. Previous to founding his company, he worked as a researcher at Citadel and Credit Suisse. Last but not least, he holds a PhD. in Finance from Indiana University.

In order to provide a quick overview of the topics covered, these are the chapters of the book:

  1. Trading Terminology
  2. Trading Costs, Alpha, and Risk
  3. Schedule Based Single-Order Algorithms
  4. Dynamic Single-Order Algorithm
  5. Oportunistic Single-Order Algorithms
  6. Practical Complications in Single-Order Algorithm Development
  7. Multi-Order Algorithms
  8. Child Order Pricing and Sizing
  9. Smart Routers
  10. Performance Measurement
  11. Performance Measurement Techniques and Issues

Machine Learning for Finance – Jannes Klaas

Despite the book having mixed reviews on Amazon, it serves as an excellent first approximation toward the machine learning world, and from a financial angle. The first couple of chapters covers the basics of machine learning in a very intuitive way. I personally found it very helpful for better understanding the basics of logistic regression, forward passes and backpropagation, and neural networks.

Unlike other books presented on this list, “Machine Learning for Finance” basic programming skills in Python, in order to leverage it and create a neural network from scratch. In later chapters, the author mostly focuses on giving a broad presentation of all the main topics or categories that are currently very active in the machine learning community.

In order to provide a quick overview of the topics covered, these are the chapters of the book:

  1. Neural Networks and Gradient-Based Optimization
  2. Applying Machine Learning to Structured Data
  3. Utilizing Computer Vision
  4. Understanding Time Series
  5. Parsing Textual Data with Natural Language Processing
  6. Using Generative Models
  7. Reinforcement Learning for Financial Markets
  8. Privacy, Debugging, and Launching your Products
  9. Fighting Bias
  10. Bayesian Inference and Probabilistic Programming

Python for Algorithmic Trading – Yves Hilpisch

This book is definitely the best recommendation for readers that know a little Python and want to leverage it toward creating automated trading strategies. Unlike other books presented here, it focuses mainly on providing simplified versions of real-world tasks that quantitative developers do on a daily basis, like developing a strategy, backtesting, and implementing it.

What I find especially interesting about this book is that it backtests shows how to backtest a simple strategy under the two main backtesting approaches: vectorized and event-driven backtesting (you can read more about both approaches in this article I wrote).

The author also covers the topic of how to deal with live data and WebSockets, which is one of the main aspects one has to cover for deploying algorithms that trade based with (relatively) low frequency. In order to stay broker-agnostic, he shows how to create a simple WebSocket server in order to use it on the client-side.

Yves Hilpsich is the founder of the Python Quants, one of the leading institutions when it comes to teaching quantitative finance and Python. He holds a degree in Financial Mathematics and definitely leverages it in order to clearly present otherwise complex topics in a way that would be otherwise difficult to understand.

The chapter on automating trading operations is also very valuable since it shows how to manage capital according to the Kelly Criterion.

You can listen to the author on the podcast “Chat with Traders”, hosted by Aaron Fifield, on the youtube video below.

In order to provide a quick overview of the topics covered, these are the chapters of the book:

  1. Python and Algorithmic Trading
  2. Python Infrastructure
  3. Working with Financial Data
  4. Mastering Vectorized Backtesting
  5. Predicting Market Movements with Machine Learning
  6. Building Classes for Event-Based Backtesting
  7. Working with Real-Time Data and Sockets
  8. CFD Trading with Oanda
  9. FX Trading with FXCM
  10. Automating Trading Operations

Algorithmic Trading – Ernie Chan

This book contains lots of valuable information and insights that only a seasoned practitioner such as Ernie Chan could provide. The author manages to convey a rigorous approach to designing, testing, and implementing sound trading strategies in an intuitive way and without the level of abstraction that other resources of similar value have.

By means of explaining different approaches towards mean reversion and momentum strategies, Ernie Chan also discusses the usual pitfalls that arise when switching from a backtesting setting to a real live trading scenario. Additionally, he goes through the key aspects of risk management, which are most often than not ignored by retail traders.

One aspect to keep in mind is that the scripts provided throughout the book are coded in Matlab, which could be a disadvantage considering that Python is nowadays the standard programming language used for data science and mid to low-frequency algorithmic trading. If you know how to code in at least one programming language, you will not find it difficult to understand and translate the code, since Matlab was designed to be easy to read.

In addition to being a successful hedge fund manager himself, Ernie Chan also holds a PhD. in Physics and has previously worked as a machine learning researcher at IBM and Morgan Stanley.

I strongly recommend listening to his interview on the podcast “Chat with Traders”, which you can find here:

In order to provide a quick overview of the topics covered, these are the chapters of the book:

  1. Backtesting and Automated Strategies
  2. The Basics of Mean Reversion
  3. Implementing Mean Reversion Strategies
  4. Mean Reversion of Stocks and ETFs
  5. Mean Reversion of Currencies and Futures
  6. Interday Momentum Strategies
  7. Intraday Momentum Strategies
  8. Risk Management

Systematic Trading – Robert Carver

“Systematic Trading” by Robert Carver is by some definitions the most important book that every trader of any type should read. The author covers each and every feature that a solid trading algorithm should incorporate. By means of clear, intuitive, and rather friendly explanations, the author conveys insights that would normally require multiple books to cover.

Newcomers to the algorithmic trading industry most often than not tend to overfit the parameters of a few technical indicators. The author starts by going through the biases that are common and relevant to keep in mind when developing trading ideas. After doing so, the book goes over the many ways in which a trading strategy can be tested, and the considerations one has to keep in mind in order to arrive at (more) reliable results.

Due to the fact that the author is a seasoned trader, he dedicates a considerable portion of the book to risk management and position sizing. These tend to be ignored by most traders and are the most common cause of bankruptcy.

In contrast to other books on the list, “Systematic Trading” requires almost no mathematics background, meaning that anyone with a high school diploma can read it from cover to cover.

In addition to holding a Master’s degree in Economics and being a seasoned trader and lecturer, Robert Carver is also a prolific writer. He is also the author of “Leveraged Trading” and “Smart Portfolios”. These books also deserve to be part of this article on their own, but I chose to only list his favorite book in order to keep this list finite.

In order to provide a quick overview of the topics covered, these are the chapters of the book:

  1. The Flawed Human Brain (Theory)
  2. Systematic Trading Rules (Theory)
  3. Fitting (Toolbox)
  4. Portfolio Allocation (Toolbox)
  5. Framework Overview (Framework)
  6. Instruments (Framework)
  7. Forecasts (Framework)
  8. Combined Forecasts (Framework)
  9. Volatility Targeting (Framework)
  10. Position Sizing (Framework)
  11. Portfolios (Framework)
  12. Speed and Size (Framework)
  13. Semi-automatic Trader (Practice)
  14. Asset Allocating Investor (Practice)
  15. Staunch Systems Trader (Practice)

A Practical Guide To Quantitative Finance Interviews – Xinfeng Zhou

This book is not only a valuable book for studying prior to an interview but also a helpful book for having around and brushing up on a few definitions and notions related to the mathematics behind quantitative trading.

Each chapter starts by introducing the required definitions and key ideas behind them that are going to be required throughout the exercises and brainteasers. The introduction to each topic is good for readers that have had previous exposure to the topic and just need to review them but is definitely insufficient for learning them for the first time.

Although having a deeper understanding of mathematics never hurts, some chapters are more relevant for sell-side quants and not so much for developing trading strategies on your own or in a proprietary firm. The topics covered in the chapters “Brain Teasers”, “Calculus and Linear Algebra”, and “Probability Theory” should be understood by any algotrader worth it’s salt.

The author holds a Ph.D. from the Massachusetts Institute of Technology (MIT), meaning that it is safe to assume that he knows a thing or two when it comes to quantitative topics. Additionally, he worked at Barclays, Goldman Sachs, and Point72, and currently works at Millenium Management. More importantly, he manages to explain each problem and exercise as if they were easy.

In order to provide a quick overview of the topics covered, these are the chapters of the book:

  1. General Principles
  2. Brain Teasers
  3. Calculus and Linear Algebra
  4. Probability Theory
  5. Stochastic Processes and Stochastic Calculus
  6. Finance
  7. Algorithms and Numerical Methods



No responses yet

Leave a Reply

Your email address will not be published. Required fields are marked *