Book Review: Machine Learning for Asset Managers

Machine learning for asset managers. 2020. Marcos M. López de Prado. Cambridge University Press (Cambridge Elements in Quantitative Finance Series).

Some asset managers see machine learning (ML) as a major advance to improve analysis and forecasting. Others argue that these technologies are just specialized tools for quantitative analysts that will not change fundamental asset management practices. Machine learning for asset managers, the first in the Cambridge Elements in Quantitative Finance Series, is a short book that does not fully answer this big question or serve as a basic text on the subject. However, it does show how the application of correct data analysis techniques can have a significant impact in solving difficult asset management problems that cannot be solved by classical statistical analysis.

Subscribe Button

The traditional approach to the broad topic of machine learning focuses on general prediction techniques and classification of supervised and unsupervised learning models by presenting differences in machine learning and deep learning, as well as general topics of artificial intelligence. (For a traditional general review, see Artificial intelligence in asset management by Söhnke M. Bartram, Jürgen Branke, and Mehrshad Motahari.) Marcos M. López de Prado, chief investment officer at True Positive Technologies and professor of practice in the Cornell University College of Engineering, uses a more modest but compelling approach to present the value of machine learning. This short work will help readers appreciate the potential power of machine learning techniques as they focus on solutions to pesky asset management problems.

López de Prado’s demonstration of problem-solving techniques provides a meaningful taste of machine learning to a wide audience. However, the book’s primary audience consists of quantitative analysts who want to read about new technologies and gain access to Python code that will begin implementing management solutions. A more in-depth analysis can be found in López de Prado’s longer work on the subject, Advances in financial machine learning.

The book’s excellent introduction explains why machine learning techniques will greatly benefit asset managers and why traditional or classical linear techniques have limitations and are often inadequate in asset management. It makes a strong case that ML is not a black box but a set of data tools that advance theory and improve data clarity. López de Prado focuses on seven complex problems or themes where the application of new technologies developed by ML specialists will add value.

A Billboard For Artificial Intelligence In Asset Management

The first major topic involves problems with covariance matrices. Noise in the covariance matrix will influence any regression or optimization analysis, so techniques that can better extract signals from noise will improve portfolio management decisions. A second topic in this same general area demonstrates how to “deconstruct” the covariance matrix by extracting the market component that often overwhelms other valuable covariance matrix information. The expansion of data signal mining techniques will better support asset management decisions.

Next, López de Prado explains how the distance matrix could be an improved way of looking beyond correlation and how the concept of entropy or codependency from information theory could be a useful tool. Building blocks, such as distance functions and clustering techniques, can account for nonlinear effects, anomalies, and outliers that can unduly influence traditional correlation analysis. For example, optimal clustering to aggregate data of similar quality can be used as an unsupervised learning technique that can effectively provide more insight into relationships across markets than is found in the traditional correlation matrix.

For those interested in the fundamental problem of forecasting, López de Prado discusses a frequently overlooked topic of financial classification–that is, setting prediction targets as a key issue in supervised learning. Horizon returns are not the only nor the best way to categorize data for predictions. For example, most traders are not interested in the difficult problem of predicting a point estimate of where a stock will be in a week or a month. However, they are most interested in a model that accurately predicts market direction. In short, the labels of what is expected are important.

Tile The Current Issue Of Financial Analysts Magazine

The book addresses the basic problem of sValues ​​and the concept of statistical significance. Interest in this topic is growing in finance due to the “zoo” of statistically significant risk premiums that cannot be replicated outside of the sample. This discussion demonstrates the broad application of ML as a general tool, not only for problem solving but also for improving theory development. ML technologies such as de-aliasing, or MDI, mean decreasing precision, or MDA, can be effective and more efficient alternatives to s-Valuable.

Since Harry Markowitz’s innovations, portfolio creation has been a constant source of frustration for asset managers. The “Markowitz curse,” which limits the successful use of optimization when it is most needed, can be addressed by using machine learning techniques such as hierarchical clustering and nested cluster optimization to tease out data relationships and simplify the optimal portfolio solution.

The final topic is tests for overfitting, which is a major problem for any quantitative asset manager trying to find that ideal. ML techniques combined with Monte Carlo simulations, which utilize fast computing power, can be used to provide multiple backtests and suggest a range of possible Sharpe ratios. A model with a high Sharpe ratio might just be a matter of luck—one return path out of wide. Using machine learning can better identify erroneous strategies and the potential for type I or type II statistical errors. Finding failures in the lab will save time and money before strategies are put into production.

Machine learning for asset managers It uses color to display better graphics and contains a great deal of Python code to help readers who want to implement the techniques presented. Code snippets are useful for readers who want to use this research, but at times, the combination of code and text in this book can be confusing. Although the author is adept at explaining complex topics, some of the steps, transitions, and conclusions are difficult to follow for anyone who lacks extensive quantitative knowledge. This work mixes some of the author’s practical research projects, but this may be a drawback for readers looking for connections between technologies in order to think about machine learning holistically.

Pioneers Of Artificial Intelligence In Investment Management

Brevity is the advantage of this work, but a longer book would better support the author’s attempt to show how machine learning can facilitate the development of new theories and complement classical statistical theories. For example, the introduction to the book provides one of the best motivations for using machine learning in asset management that I’ve read. In just a few short pages, he addresses common misconceptions, answers frequently asked questions, and explains how machine learning can be applied directly to portfolio management. López de Prado has practical insights that most technical writers lack, so drawing broadly on his deep knowledge of the science of machine learning will be beneficial to readers.

In summary, Machine learning for asset managers It successfully shows the power of ML techniques to solve difficult asset management problems, but it should not be seen as an introduction to the topic for general asset managers. However, learning how these techniques can solve problems, as explained by the highly successful asset management author, is well worth the book’s modest price.

If you liked this post, don’t forget to subscribe Venture investor.

All posts are the opinion of the author. As such, it should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of the CFA Institute or the author’s employer.

Professional learning for CFA Institute members

CFA Institute members are empowered to report self-earned and self-report Professional Learning (PL) credits, including content on Venture investor. Members can easily score credits using their online PL tracker.

Home Page




Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors