Best Resources to Learn Machine Learning and Apply It to Finance: Books, Courses, and YouTube (2019)

3 min read

In today’s technologically advanced world, we’re starting to turn to the machines we create to aid us. The algorithms designed by data/AI scientists are becoming ever so powerful — and now they’re showing their prowess. Much like Google’s Deepmind AI reinvented the way chess should be played, machine learning applied to finance can provide valuable insight into how money should be made.

Having covered the theory of machine learning in finance, we now dive in deeper and ask the following question: “Well, how can we actually improve our knowledge and understanding of machine learning — and apply it to be one step ahead of the competition?” This question is a pressing issue for both IT and finance professionals — so we’ve compiled an exhaustive list of the best resources to learn machine learning and apply it to finance.


Courses can be considered to be books on steroids: they offer structured material created by industry professionals, but their real power comes with all the additional material they serve: on-demand video, interactive content,

Machine Learning for Finance in Python

Price: First chapter for free, the other chapters require to upgrade your DataCamp profile ($25 per month)

Authored by Nathan George, Assistant Professor of Data Science at Regis University, this course thoroughly explores time series data, how stocks perform historically — and how this data can be used to create an accurate and profitable stock trading strategy.

The course is divided into four chapters:

  1. Preparing data and a linear model. This introductory chapter provides an overview of how machine learning can be used in finance and prepared for machine learning algorithms.
  2. Machine learning tree methods. In this chapter, the magic of predicting the future comes to life: tree- and forest-based machine learning methods can try and predict future values of stocks’ prices, so now it’s time to learn them.
  3. Neural networks and KNN (k-nearest neighbors algorithm). This chapter explores how data can be normalized and scaled for later use in neural networks — this is another method of predicting future values of stocks’ prices.
  4. Machine learning with modern portfolio theory. This chapter features Modern Portfolio Theory: using rules like Sharpe ratio, you will find and predict the best portfolios.

Machine Learning for Trading

Price: Free

Georgia Institute of Technology’s brainchild, the course will teach you how to implement machine learning — starting from information gathering to the final model — to create a trading strategy. In order to do this, probabilistic machine learning approaches will be used, utilizing linear regression, KNN and regression trees. This course is part of the Artificial Intelligence for Trading which covers a wider area.

The course is divided into three lessons:

  1. Manipulating Financial Data in Python.
  2. Computational Investing.
  3. Machine Learning Algorithms for Trading.

Upon completion, you will have these competencies:

  • Understanding data structures, data mining, and machine learning algorithms.
  • Constructing a stock trading software system.


Although it may be tempting to disregard the importance of books (didn’t we just say that “courses are books on steroids”, so who needs books?), their major advantage lies in the fundamentalness: as machine learning relies heavily on advanced mathematics, these books are tried and tested tools of helping you understand machine learning thoroughly.

Advances in Financial Machine Learning

In this book, noted financial scholar Marcos Lopez de Prado will provide you with a foundational understanding of the “machine learning + finance” duo: structuring big data, researching to find the best machine learning algorithms, backtesting and cross-checking your findings, and applying them in real-life scenarios.

Hands-On Machine Learning with Scikit-Learn and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems

Although not strictly finance-related, this is an invaluable resource for understanding the bigger picture of machine learning. This book doesn’t require you to dive deep in mathematics; on the contrary, it utilizes concrete examples to teach its reader. The highlight of this book is the knowledge of working with scikit-learn and TensorFlow — two production-ready Python frameworks that make the power of machine learning truly shine. Additionally, a newer version of this book is set to release on July 5, 2019.


Nowadays, YouTube is also an impressive education platform: countless creators are all putting maximum effort into creating valuable and insightful content. Moreover, they’re doing it for free — but you can always support your favorite YouTube teachers via a sponsoring service like Patreon. Another advantage of YouTube courses is the speed at which they can be created: in a fast-paced area like data science, changes happen blazingly fast — and YouTube creators can react quickly.

An incredibly talented Python programmer sentdex has been teaching Python on YouTube for years, amassing numerous “Thank you for this video, I finally understand Python” comments. As Python is an ideal tool for machine learning/AI, Sentdex covers these areas as well: his Machine learning with Python and Python programming for finance courses provide great material that balances extensive theoretical background with hands-on experience.

Machine learning and data science wizard Siraj Raval’s channel is another great resource: his AI for business and Machine learning journey courses are equally awesome.

Advice on using machine learning in finance

It’s always nice to decorate your walls with course completion certificates, but which actionable advice can we get from these books and courses?

Set business KPIs: This will help you define the estimates realistically and avoid budget draining. How can this goal be achieved? This is where data science-oriented professionals step in: they can validate the idea and formulate appropriate KPIs.

Avoid overcomplication: It may be tempting to try to improve every single aspect of the business; however, the efforts would be too spread out. A financial organization (e.g. a bank) can focus on a single important feature: in a bank’s case, this can be the response time when dealing with client requests.

Don’t ignore other areas of data science: Although machine learning is a powerful tool in and of itself, data science subsets like statistics, data visualization, and data engineering help to manage your data in an efficient manner. Indeed, they aren’t as flashy as machine learning or artificial intelligence — but they are the backbone of a successful data strategy.


Machine learning may seem like a daunting and mysterious area — after all, it’s a buzzword which has been overused by the media to talk about AI. However, machine learning isn’t some black magic — with just enough patience and perseverance, you can become proficient in this area and crush your competition!

Denis Kryukov Denis Kryukov is an author at Soshace, an online hiring platform that connects IT professionals and companies.

Leave a Reply

Your email address will not be published.