Current times are indeed turbulent again, and for better or worse, financial markets are displaying their high potential to react to these changes on a global scale. Given this new reality, when it comes to data, itʼs increasingly valuable to be able to identify and process new information, spot emerging relevant topics, and assess their potential impact on financial markets and the economy.
To achieve this, investors are progressively more interested in gaining real-time access to a broad range of datasets. Much of this data, once accessed, has to be processed and aggregated in order to be interpreted on reasonable time scales with coherent metrics. This makes tools such as Machine Learning and Natural Language Processing techniques powerful and handy for the skilled investor.
Brain uses both to help the world make sense of its data, and amidst the COVID-19 crisis, put these techniques to work to analyze a range datasets.
Introducing Brain and its Approach
Brain is a research company that develops proprietary signals and algorithms for investment strategies using Machine Learning (ML) and Natural Language Processing (NLP) tools. The company has developed a scientific and rigorous approach based on its foundersʼ years of research and experience in implementing statistical models as well as state-of-the-art software.
Brainʼs proprietary NLP platform automatically scans and analyzes around 2,000 financial news sources in near real-time, extracting trends and sentiments surrounding the topics discussed. This can be used to detect the rise of new topics as well as trends and common sentiments across multiple news source.
Brainʼs NLP platform can be applied to monitor financial sentiment at market level (with the Brain Market Sentiment), at company level (with the Brain Sentiment Indicator, or BSI), and also at topic level – such as COVID-19, or oil. Brain used all three levels to analyze recent financial news.
Findings from Brain
Financial News and the COVID-19 Outbreak
One example of Brainʼs NLP platform and its ability to monitor current market trends is its analysis of recent news and topics in financial media related to COVID-19. The flow of information through new sources can be measured across various dimensions:
The General Equity Market
Snapshot of January – March 2020:
Negative news surrounding financial markets already began mounting almost a month before the market crash. This is shown by charting the moving averages of the Brain Market Sentiment over time, which analyzes financial news sentiment at the market level. We see a crossing of the 30-day moving average over the 90-day moving average as soon as the end of January.
Snapshot of January – May 2020:
After the moving averages crossed at the end of January ahead of the February-March market crash, positive financial news slowly increased throughout April, with the 30-day moving average crossing back over the 90-day moving average at the end of the month.
The rise and plateau of news topics
In the weeks leading up to March, the peak of financial news discussing COVID-19 occurred January 26-27, 2020. Meanwhile, the first significant presence of news discussing COVID-19 in financial media was detected on January 20, 2020, just 10 days before the W.H.O. declared a global health emergency.
Brain groups news by financial topic. This helps you analyze what percentage of news topics discussed COVID-19, in addition to simply the percentage of overall news reports. From that angle, the percentage of new topics discussing COVID-19 peaked early on January 24, followed by a second greater spike on January 27.
One takeaway is that in its early months, COVID-19 took up a relatively small percentage of financial news stories (less than 0.5% at peak times), but had a relevancy of broader reach, impacting up to 20% of financial news topics. A second takeaway is that this trend spiked relatively early on in January.
Impact on individual stocks
Brain also tracked news sentiment fluctuations surrounding individual companies – airlines, for instance. The Brain Sentiment Indicator abruptly displayed negative sentiment for Chinese airlines like China Southern Airlines around January 20. This was around the same time countries outside of China, like South Korea and Japan, began reporting their first cases.
Following a rise of cancellations in flights to China, the same trend impacted other global airlines just a few days later on January 30. This was the same day the W.H.O. declared a global health emergency. Take a look at American Airlines as an example.
Financial News and the 2020 Oil Crash
Brain can help give insights into news trends about other macro events, too. Take a look at analysis of financial news surrounding the recent crash in oil prices.
The 30-day moving average of the Brain Sentiment Indicator on oil sharply crossed the 90-day moving average at the end of January. Negative news sentiment surrounding the oil market began mounting over a month before the sell-off in March 2020.
Tracking News Sentiments for Individual Companies
The Brain Sentiment Indicator can also be a great tool for measuring and monitoring financial news sentiment amidst events specific to individual companies. Check out Brainʼs data for Boeing, Tesla, and Facebook in the recent years.
On October 29, 2018, a Boeing 737 Max aircraft crashed shortly after take-off. This was followed a second accident on March 10, 2019, which intensified investigations into Boeing and led the U.S. FAA to temporarily ground all Boeing 737 Max 8 and 9 aircraft.
With the Brain Sentiment Indicator, we see a temporary dip in October that recovers within a few weeks, followed by a more severe, much more long-term fall.
Between April and May 2019, the Brain Sentiment Indicator for Tesla was largely negative. This trend took place as an increased flow of news discussed possible escalations in trade tensions with China – which would potentially factor into demand for Teslaʼs cars.
The Brain Sentiment Indicator can also be used to understand trends surrounding Facebook throughout early 2018. Interestingly, the Brain Sentiment Indicator already detected a rising amount of negative news concerning Facebook at the end of February 2018, a couple of weeks before the Cambridge Analytica scandal occurred
and Facebook saw a large drop in its stock price. While the increase in negative sentiment alone does not necessarily mean that a stock price is always going to fall, it can help raise red flags and provide clues in estimating a stock priceʼs future performance.
Continue your analysis using Brainʼs datasets on IEX Cloud
Want to see more? Continue exploring trends in financial news sentiments by accessing the data directly. Brain provides these datasets – alongside several others – on IEX Cloud as a Premium Data offering. See a full list of the platformʼs endpoints and datasets offered by Brain.
Premium Data is available as an add-on to all paid plans on IEX Cloud, and provides access to curated, third-party data through a single, easy-to-use API.
More about Brain and its data
Brain, based in Milan, Italy, is a research company that develops proprietary signals and algorithms for investment strategies using Machine Learning (ML) and Natural Language Processing (NLP) tools.
Brainʼs market proposition is two-fold: first, financial firms can combine Brainʼs proprietary datasets with their own strategies to create an investment model tailored to their needs. Alternatively, clients can obtain consultancy support to use ML, NLP, or other advanced techniques to enhance their investment approach with a wider range of data and methodologies.
While the company offers a range of services, its datasets are a good illustration of its general approach: a few examples are the Brain Sentiment Indicator (BSI), as seen above, as well as the Brain Machine Learning Stock Ranking (BSR) and the Brain Language Metrics on Company Filings (BLMCF). Brain also supports clients in the development of tailored strategies.
The Brain Machine Learning Stock Ranking (BSR) is a dataset that can be used for quantitative long and long/short strategies. Itʼs based on a combination of non-linear Machine Learning classifiers and aims to estimate the future return of approximately 1,000 U.S. stocks. The model combines stock-specific market data, such as fundamental ratios, price evolutions, general market data, and calendar anomalies. The BSR uses a dynamic universe that is updated every year to avoid survivor bias, with various prediction time horizons available (2, 3, 5, 10, and 21 working days).
The Brain Language Metrics on Company Filings (BLMCF) is a more recently developed dataset, with the objective of monitoring several language metrics on 10-Ks and 10-Qs company reports for approximately the largest 1,000 U.S. stocks.
Matteo Campellone, Ph.D, MBA
BRAIN – Head of Research
Francesco Cricchio, Ph.D
BRAIN – Chief Technology Officer
Building with Financial Data Sets and APIs
In this article, you'll learn how to incorporate financial data into your application. First, you'll learn more about how to plan for and use real-time financial data in your applications. Then, you'll use the financial API provided by IEX Cloud's Apperate platform to build a financial stock website using NestJS.
Taming the Chaos: Sourcing Financial Data from Multiple Providers
Combining multiple data sources can help your app or website stand apart from the pack. However, it also comes with several challenges. This guide can get you started on the right foot.
12 Questions to Ask Your Financial Data Provider Before You Launch Your App
Creating a fintech app or website that uses financial data? Need a stock market API? Before jumping in, you need to know what you're getting into with your financial data or stock market data provider. We've put together a list of 12 questions to help you navigate potential challenges and identify key success factors.