Duke University & University of Luxembourg
| September 07, 2016
Brandt & Gao (Duke University / University of Luxembourg) use RavenPack's Big Data analytics to predict oil prices.
News about macroeconomic fundamentals and geopolitical events affect crude oil markets differently.
Oil prices are hard to explain and predict. Hamilton (2008) suggests that the real price of oil follows a random walk without drift. Kilian and Baumeister (2014), who explore an exhaustive set of oil pricing factors compiled from the literature, conclude that the
explanatory power of these factors vary over time and that different factors are important at different time horizons.
The ever changing nature of this predictive relationship contributes to the difficulty of forecasting oil prices. Furthermore, oil prices are not only related to economic fundamentals but also to geopolitical events that are much harder to quantify. For example, policy issuance is a one-time event for which it is impossible to set up a time series record to then quantitatively relate this event to oil prices.
News analytics provide a way to quantify both macroeconomic and geopolitical events. It not only offers timely analysis of the news content but also captures both the raw information as well as the market perception of the news. We use news analytics to compare and contrast the
importance of macroeconomic and geopolitical information for crude oil
We consider a broad cross-section of macroeconomic and geopolitical news. The goal of the paper is to investigate the role of these different types of news for both
crude oil prices and trading
activity in oil markets. We rely on news sentiment scores provided by RavenPack to capture the sign, magnitude, relevance, and novelty of the news.
Our results highlight important differences between the roles of macroeconomic and geopolitical news as well as more subtly differences between news from oil-producing versus oil-consuming countries. News about economic growth is the strongest predictor of oil returns over coming months, confirming the findings of Kilian (2009) that macroeconomic growth generates demand for oil.
The gradual diffusion of public information and resulting momentum and predictability in oil returns can be attributed either to behavioral reasons such as investor inattention or to the interaction of different types of traders who only trade based on news or news-based price movements (Hong and Stein, 1999). The response of oil prices to geopolitical events is even stronger and immediate. News is incorporated into the oil price at once without follow-on momentum, consistent with informational efficiency.
Please use your business email. If you don't have one, please email us at email@example.com.
We will process your personal data with the purpose of managing your personal account on
RavenPack and offering our services. You can exercise your rights of access, rectification,
erasure, restriction of processing, data portability and objection by emailing us at firstname.lastname@example.org. For more information, you can
Your request has been recorded and a team member will be in touch soon.
High inflation has returned in developed markets after decades of lying low. In our latest paper, we show how to build an inflation-based asset allocation strategy using sentiment data and we illustrate that sentiment-based strategies outperform models that depend merely on past observed inflation values.
This year's RavenPack Research Symposium brought two intense days of knowledge sharing in London and New York, from 25 top experts in natural language processing, quantitative investing and machine learning. Together, we explored how firms can leverage new language models to generate alpha, better manage risk and respond to calls for more sustainable investment practices.
Human capital is at the heart of value creation. Our latest research demonstrates how unprecedented workforce insights, sourced from over 200 million job postings, can generate more alpha.