Whitepaper: Predicting Crude Oil Prices

Using Big Data Analytics & Machine Learning Algos

In this paper (released in early Jan 2016), we answer the question – Where are Brent Crude prices headed? This is a question that continues to baffle most of us despite the significant drop in Crude oil price in the last year (2015). Several theories and projections are in place already, some predicting sub-$20 prices through the year 2016, while others expecting a moderate recovery ! Most predictions are based on technical / fundamental analysis of oil and use analyst outlook to further support the projections. This paper uses a unique approach of modelling prices and fundamental data by using Artificial Intelligence / Machine Learning based techniques. It uncovers the hidden relationships between different variables affecting the oil prices, and also uses the model for predicting Brent crude prices.

How does it work?

  • We used month-end prices and fundamental data over the last 14 years
  • Key Fundamentals most likely to be associated with Crude Oil Price Discovery were chosen
  • Fundamental data includes Real GDP of OECD & Non-OECD Countries, Crude Production and disruptions in key geographical areas, Supply and consumption patterns in major countries, Spare Crude Production capacity and OECD Inventory levels, etc.
  • Decision forests with multiple regression techniques were used for building the model

What can you learn from this paper?

The paper is an abridged version of our complete Research paper. So while you might have to research on some terms and concepts that we have used (especially if you are not familiar with Machine learning techniques), the paper gives in brief the various stages of a machine learning based prediction model. Thus, we expect this paper to intrigue you, and while it provides you a glimpse into what Big Data Analytics and Machine Learning can do for you, it will also leave you with several interesting questions.

  • Understanding the relationships between several fundamental factors that affect crude oil prices.
  • How to visualize relationships between various factors. Should only factors with good correlations be taken into the model?
  • Which factors are more important than others when it comes to price discovery of Crude Oil?
  • How to combine fundamental and price data to predict crude oil (or any other commodity, for that matter) prices (or demand / supply / yield)?

Now, it’s also a web-application !

In 2017, we released a web-application based on this whitepaper, so our readers can play around with the advanced visualizations and build their own models to predict the prices up to 18 months ahead. This solution is updated regularly and so has latest data to get the most relevant predictions.

When you download this whitepaper, you also receive a link to the web-application !

You can download this paper for Free by submitting the form on the right. If you are an existing subscriber from an Energy / Commodity company, you don’t need to fill this form again, you’ll receive this paper in your registered email id !

Please note that our Publications are shared selectively only with people from Energy / Commodity / Banking / Financial Services companies. We frequently get requests from generic domains (like gmail, yahoo, etc.) and sometimes, despite our best efforts to verify their identities, we are unable to do so. Such requests unfortunately have to be declined. So we would request you to use your official email id for downloading our publications – and be assured that you’re in good company !

Download Requests from vendors / consultants are also taken up separately, and approved only if there is a viable business case for doing so. Please write in separately to us on info@riskedgesolutions.com – we read and respond to every personalized email !