by Eugenio Ciamprone
In 2010, studies conducted by universities in the US, such as MIT and Indiana University, showed that big data analysis and the monitoring of conversations on social networks enabled traders to earn more than those who did not have this type of information at their disposal.
A few years later, in 2013, Twitter introduced the possibility of following the conversations of companies listed on Wall Street through the use of Ticker (abbreviations that identify companies that are listed on financial markets) and CashTag where the share symbol of the company is preceded by the sign of the American dollar.
Eight years on, it can be argued that finance and investment are increasingly running on the keys of our smartphones and PCs that feed conversations on the web. This is also thanks to the entrepreneurs whose tweets have helped guide investors’ choices in recent years, and who have succeeded in conveying more than just a good quarterly report.
One example of this is Tesla, whose Nasdaq-listed stock has risen more than 500% in the past 12 months, thanks more to its futuristic concept of the electric car than to its actual sales results. Tesla CEO and SpaceX founder Elon Musk is now considered a financial guru, and each of his tweets moves millions of dollars.
Another sector that has been strongly influenced by social media and has seen explosive growth in recent years is cryptocurrency. Here again, following sentiment and influencers in the sector helps to achieve better performance on assets that are still very volatile, driven by Bitcoin and the vision of a new financial paradigm.
Returning to the equity sector, the fact that monitoring online conversations is now necessary for better investment performance is also shown by the returns of the Buzz NextGen AI US Sentiment Leaders index: from January 2020 to January 2021 it gained more than 90%, almost five times the S&P500. The index is made up of a basket of 75 stocks that changes every month on the basis of buzz, i.e. the buzz that artificial intelligence algorithms continually intercept on the web regarding sentiment and discussions about the stocks themselves. Stocks, therefore, are included in the basket on the basis of interest recorded on the net.
What are the types of data that must be taken into account to intercept sentiment on financial assets?
Extracting reliable information from the web is an increasingly complex task, and artificial intelligence algorithms have also become more efficient over time. The examples shown above demonstrate how the use of alternative data sources to the use of corporate information alone allows us to intercept investor sentiment and behaviour with greater speed and efficiency.
Alternative Data is data that does not originate from the corporate context, but is found in the form of text, images or video, online or on social networks.
In the field of investment, the processing of large volumes of alternative data allows signals and trends on specific assets to be intercepted in advance. The more information analyzed, the better the performance. The evaluation of investments is built on the ‘digital footprints’ present on the Net, whose predictive capacity is now superior to traditional data analysis.
The use of alternative data also makes it possible to identify so-called weak signals, information that is apparently random or initially not very relevant, but which, when put together, can represent significant indications.