Online posts about ESG, which stands for Environmental, Social, and Governance, influence the trading of stocks in a major way.
LONDON, UK, July 12, 2019 /EINPresswire.com/ — As part of a syndicated social intelligence report, DigitalMR harvested 4.5 million online posts about 11 major banks from Twitter, News, Blogs, Forums, Reviews and Videos. All relevant posts were annotated with topics and sentiment using custom machine learning models.
The daily stock price of the banks involved was then regressed against various time series of variables from the annotated posts.
Here are some specific examples of correlation between ESG posts and bank valuations:
Negative posts about ESG from all online sources were correlated against the DB stock price using a 30 day rolling average. The result was negative correlation as one would expect R2 = -0.40.
Looking at posts about ESG coming from news only i.e. editorial content, regardless of sentiment, the correlation factor of monthly total posts and monthly valuation was R2 = 0.79.
Royal Bank of Scotland
For RBS the correlation factor was even higher when using News posts with positive and neutral sentiment, at R2 = 0.87. In this case the 30 day rolling average was used for both variables.
Using the same parameters as for the RBS case, only with posts coming from all sources rather than just news, the correlation was a staggering R2 = 0.92.
Out of the 3 elements of ESG, Governance is the strongest, with the largest number of posts. The direction of causation is easily implied; traders (human or algos) read the news on bank governance which influence sell or buy decisions.
For more findings from the Social Intelligence for Banks report, contact DigitalMR.
Sophia Papagregoriou, Sales & Marketing Manager of DigitalMR said: “We started this initiative not knowing what would be the outcome, and it is proving to be increasingly exciting, with results such as the above! This is definitely the right time for banks to invest in social intelligence, to assess their positioning and improve their performance.”
DigitalMR is a London based technology company with proprietary solutions for social intelligence (listening & analytics) – listening247, and DIY private online communities – communities247. It specialises in the use of Artificial Intelligence for high accuracy text and image analytics and the integration of Social, Survey, and Sales data for unique actionable insights that could not otherwise be obtained. listening247 was developed specifically for market research, offering the highest possible sentiment and semantic (topics) accuracy in any language. It can be used to analyse data on any topic or product category, from any country around the world, on all social media and other online sources such as blogs, forums, news and reviews. DigitalMR has a variety of current and past clients and partners, and works with multinationals in various sectors such as FMCG, retail, banking & finance, telecoms, and NGOs.
Source: EIN Presswire