Learning Resources for Machine Learning in Quantitative Finance

Summary

The document recommends resources for learning how to apply machine learning and deep learning to finance. It distinguishes between financial research practice, neural-network implementation, and data-science programming. The central recommendation is a finance-focused machine-learning text that covers challenges particular to financial data, including data handling, strategy research, backtesting, labeling, bagging, and meta-models. For hands-on deep learning, it points to a practical introduction using Scikit-Learn, Keras, and TensorFlow, while a Scala resource is suggested for data-science programming and production-oriented development.

The recommendations are framed as one respondent’s preferences, not as a comparative review or a tested curriculum. The answer gives no discussion of mathematical prerequisites, the relative merits of alternative books, or how to validate machine-learning strategies against overfitting and market changes. It is most useful as a starting shortlist for a reader with some programming and quantitative background, rather than as a complete study plan.

Key ideas

• A finance-specific machine-learning resource can address research and data challenges beyond model mechanics.
• Topics highlighted include data handling, strategy research, backtesting, labeling, bagging, and meta-models.
• A practical deep-learning introduction can provide hands-on work with neural networks and common frameworks.
• Production-oriented data-science programming is a separate learning need from financial modeling.
• The list reflects personal recommendations and does not compare resources or establish a full curriculum.

Tags

# Reference request for AI / deep learning for finance


# Reference request for AI / deep learning for finance












Are there any particular references that are recommended for learning about AI / deep learning and how to apply them to quant finance?

I have written programs that price options, done sentiment analysis, SVM etc (to give an indication of my programming and mathematical background).

Thanks.

## Answer by StackG (score 3)

https://quant.stackexchange.com/a/58232

My all-round favourite for ML-in-Finance: "Advances in Financial Machine Learning" by M. de Prado (https://www.wiley.com/en-hk/Advances+in+Financial+Machine+Learning-p-9781119482086)

He doesn't spend so much time dealing with specific ML models, but talks about the unique challenges faced by financial data scientists and the ways to handle data, conduct strategy research and backtesting, and the style and techniques (labelling, bagging, metamodels...) to apply generic ML techniques in a financial setting.

My favourite introduction to Deep Learning: Hands-On Machine Learning with SciKit-Learn and TensorFlow

Intensive introduction to Neural Nets via Keras (the author created this library) and Tensorflow, with lots of Python code examples

My favourite pure-programming for Data Science-type problems: "Scala for Data Science" by P Bugnion

An immersive introduction to Scala with a focus on Data Science problems, and a focus on how to move products into production safely

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.