Advanced Positioning, Flow, and Sentiment Analysis in Commodity Markets, 2nd Edition

CHAPTER 14Machine Learning – A Machine's Perspective on Positioning

Chapter objectives

• 14.1 Introduction to Machine Learning (ML)
• 14.2 Decision Trees
• 14.3 Feature Importance
• 14.4 Random Forests
• 14.5 Using ML to Trade

In this chapter, decision trees and random forests are introduced as ways of uncovering relationships between changes in positioning and changes in commodity prices.

Tree‐based learning algorithms, which include decision trees and random forests, are amongst the most‐used machine learning methods. The objective is to identify which aspects of positioning are the most useful in helping to understand commodity markets from a machine's perspective.

Machine learning applied to positioning data to predict prices is particularly useful in the analysis of positioning data, with ‘feature importance’ a powerful way of identifying new patterns and new relationships in positioning.

These are insights that can help improve how other positioning signals, indicators, and models are interpreted and used.

14.1 Introduction to Machine Learning (ML)

Machine learning (ML) is the scientific study of algorithms and statistical models that computer systems use to effectively perform a specific task without using explicit instructions, relying on patterns and inference instead. It is seen as a subset of AI.

Machine‐learning algorithms build a mathematical model of sample data, known as ‘training data’, in order to make predictions or decisions without being explicitly programmed ...