JB Marwood - Guest Blogger
| June 16, 2017
Artificial intelligence (AI), particularly the use of natural language processing (NLP) and sentiment analysis on news, continue to be a fertile ground for global macro investors seeking out new ways to generate alpha.
A recent study from
J. P. Morgan
showed that global equity portfolios can be successfully constructed using economic sentiment indicators. Meanwhile,
the Handbook of Sentiment Analysis in Finance
provides a compendium of such data driven research.
In this post, we will see that current global news sentiment is portraying a moderately positive outlook overall. However, there are enough pockets of risk to give investors cause for concern. You can request the full case study
As we shall see, a top-down analysis of global news sentiment can be an aid to portfolio managers seeking to deploy capital across geographic locations. It can also help in the task of visualising and managing risk.
A top-down look at global economic sentiment can be easily achieved using the predefined ‘Economic Outlook’ dashboard available on the RavenPack website. There it is possible to get an overview of economic sentiment around the globe and drill down further to see the news stories impacting those sentiment scores.
Sentiment scores are calculated by an algorithm that generates a range for each event between -1.00 to +1.00 where 0 indicates neutral sentiment, positive values indicate positive sentiment, and negative values indicate negative sentiment.
As you will see from the visualised Treemap below, global economic sentiment is mostly positive albeit with some sharp pockets of negative sentiment shown in red. This is based on a monthly snapshot of news coverage.
Clearly, the largest contributor of negative sentiment stems from news articles relating to the US Federal Reserve. A closer look indicates a negative sentiment score of -0.76 based on sampling 1,662 relevant news articles.
Fortunately, it doesn’t take an investment genius to understand why this is so. With the Federal Reserve expected to raise interest rates this week there has been extensive news coverage throughout the financial press.
Such a move is generally seen as a negative for economic growth (and equity markets) which is why the sentiment is shown so clearly red in the visualisation.
Although interest rate decisions are instructive of future equity movements, they only tell one part of the story, however, and a look at other entities within the sentiment dashboard reveals plenty of positive news coverage for many parts of the world.
US, Eurozone, Canada, Japan, Germany, Russia, France, UK – all these entities show positive news sentiment scores where they relate to economic factors.
In some cases this is surprising. The UK, for example, manages a positive score despite the most recent political developments that have led to a hung parliament.
Other scores are less surprising. Canada scores a high 0.62 that is consistent with recent GDP numbers out of the country and latest numbers on housing.
Japan also scores well, partly owing to recent figures from the government that show an improvement in balance of payments.
Germany also scores positively with 26 news articles cited that identify positive sentiment related to improved guidance on economic growth.
Despite the overall positive outlook there are many areas of red that might warrant investor concern.
Australia is a good example. After 25 years without recession you would think the country might score positively on economic measures. However, recent news sentiment has been mostly negative on economic grounds resulting in a monthly sentiment score of -0.33.
Stories relating to economic growth guidance, recession guidance and a drop in consumer confidence have all contributed to the negative score with one news article claiming that the country will slip into recession unless it’s energy policies are resolved.
Other areas where investors might want to tread carefully include China, Greece and Kenya.
Global news coverage and sentiment scores are able to highlight opportunities and risks that macro investors ought to consider and these might help in the formulation of an investment plan.
Sentiment around Federal Reserve interest rates, for example, suggests that investors should focus on assets that will benefit from higher yields.
Meanwhile, negative sentiment in areas like Australia suggests investors might avoid this market or even look to go short.
There are different ways to interpret such data and different ways to formulate an investment plan and this is discussed further in the accompanying full case study which you can request here:
Global Opportunities Based On News Sentiment Analysis
JB Marwood is an independent trader and writer specialising in mechanical trading systems and market analysis. He can be reached at
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.