Financial assets selection under the Factor Investing scheme with the support of a Machine Learning algorithm

Resumo

In this article, the author examines one of the most viable Machine Learning methods for its expeditious implementation in the financial industry, which does not require high computing capacity as required by the most sophisticated Artificial Intelligence techniques that have emerged today. This is the binary tree classification method, which is a supervised learning algorithm, part of the arsenal of Machine Learning algorithms.

This algorithm can be applied efficiently with a medium computing capacity and adequate knowledge of Machine Learning techniques.

An appropriate understanding of the behavior of financial markets is essential for its good implementation, particularly of the reports issued by publicly traded companies.

The case study of this algorithm in the financial industry is in a modern asset allocation scheme known as Factor Investing.

The implementation of Binary Trees for this task is suitable, as it is an asset classification task.

This study shows that through machine learning, specifically binary trees, investors and portfolio managers can identify factors that support their investment strategies.