Using Machine Learning to Identify Drivers of Oil Price

Machine learning is a subset of artificial intelligence that deals with the creation of algorithms that can learn from, and make predictions on, data. These algorithms are able to automatically improve given more data.

Machine learning is a powerful tool that can be used for a variety of tasks, such as detecting fraud, making financial predictions, and automating investment decisions. In the world of finance, machine learning is being used in a number of different ways.

For example, machine learning is being used to develop better credit scoring models. These models are used to assess an individual’s creditworthiness and are important in determining whether or not a person will be approved for a loan.

Machine learning is also being used to create algorithms that can trade stocks and other financial instruments automatically. These algorithms are able to make decisions based on data and can execute trades in a fraction of a second.

Reference [1] utilized a random forest model based on 1000 regression trees to identify different economic and financial factors that impact the price of oil. It identified the interest rates, the dollar, and the VIX index as the important drivers of the oil price.

All in all, the analysis finds strong relative performance of random forests in predicting monthly oil price changes. The performance of this purely data driven method, relative to more traditional linear methods, is likely driven by its ability to accommodate nonlinear effects and interactions between explanatory variables.

The model which was presented relies on much more timely data inputs (at daily frequency). This is particularly important, as the model promptly incorporates changes that may be occurring in financial markets. The large role of these factors in reducing prediction errors only underscores the importance of interpreting spot crude oil as an asset price.

In short, a pure data-driven method was successfully developed to identify economic and financial factors that impact the price of oil.

References

[1] Emanuel Kohlscheen, Quantifying the Role of Interest Rates, the Dollar and Covid in Oil Prices, https://doi.org/10.48550/arXiv.2208.14254