December 18, 2020
Summary of 10 of the most interesting data insights we drew from
news analytics and sentiment data in
Events such as COVID-19 and the U.S. presidential elections have
made 2020 a year unlike any other.
Below is a summary of 10 of the most interesting data insights
we drew from news analytics and sentiment
data in 2020.
Analyzing the relationship
between levels of news about “Anger” and different candidates’
winning chances revealed that the more
“Anger” was covered in the news the more Trump’s projected
chances of winning the U.S. election slipped.
The opposite was true of Biden where mentions of “Anger”
forewarned of improving winning chances.
When companies change the dates of their official earnings
releases it is often thought to be because
they want to delay the release of bad news or bring forward the
release of good news.
A portfolio investment strategy that
put this theory to the test
and added news sentiment into the mix produced solid annual
excess returns of 8.4 - 10.5%, for mid to
large-cap stocks and 18.6 - 26.7% for small-caps.
There was no evidence the U.S. presidential election was a
‘referendum’ on the administration’s response
to the Covid-19 pandemic.
Although a link between the two was noted in the Spring, Trump’s
projections were largely unaffected
by Covid news during the months of June and between mid-August
and early October. Polls also seem to
have been unmoved by mentions of the virus.
According to our media monitor, Biden’s sentiment consistently
remained above Trump’s from March 2020
till election day.
On the day before the election, our model was forecasting Biden
to win by 313 electoral college votes
(ECV) to 225. In the end, Biden won by 306 to Trump’s 232. The
was only 7 ECVs off the actual result.
When we tested to see if our news sentiment data could enhance
more traditional factors in the global FX
market, such as the carry, value and momentum we found the
resulting model was
3 times more effective
than that which was based on traditional factors alone.
A media exposure gauge using our data
both major UK lockdown periods, starting with the Leicester
lockdown in June and then the tougher
lockdown in November.
A measure of the volume of news about Covid, the Media Hype
Index, and another index - the Panic Index -
that measured the level of concern in the media,
provided interpretable warning signals
of the various peaks of the coronavirus pandemic as well as the
March stock market trough.
Research from earlier in the year
uncovered a close connection
between the number of times a BioPharma company was co-mentioned
with COVID-19 in the news and changes
to its share price.
The conclusion was that BioPharma stocks with “a higher
co-mention volume with the virus were also the
top performers in their sector,” wrote Peter Hafez, Chief Data
Scientist, at RavenPack.
News about how company executives trade their shares, also known
as insider transactions data, can
provide investors an information edge and could have
forewarned of the major market turns
caused by Covid.
A trading strategy based on insider insights delivered, “a
robust global portfolio performance with
little exposure to systematic risk factors,” says Peter Hafez,
Chief Data Scientist at RavenPack
A growing corpus of research shows that news analytics can help
forecast central bank policy
moves by analyzing relevant news in the run-up to scheduled
undertaken by researchers at the Bank of International
Settlements (BIS) found news analytics
outperformed a leading economist’s survey at predicting interest
rate changes by the Bank of Indonesia
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.