• 12 February 2021

Traditional financial data are no longer enough to make investment decisions

Traditional financial data are no longer enough to make investment decisions

Traditional financial data are no longer enough to make investment decisions 1024 683 FinScience

written by Alessio Garzonio

The GameStop case is a surprise only for those who adopted the “ostrich” strategy and therefore ignored the process of innovation and democratization that has impacted the world of investments in recent years. The force of retail investors, who express themselves, share information and organize themselves on social platforms, is present and influencing the market more and more and now it is clear that no one can ignore it anymore. Even these days, now that the focus on Gamestop stock is over, the movements of private investors are continuing to have a strong impact on the market, as evidenced by the growth of stocks related to cannabis, silver or the ride of cryptocurrencies.

The last year and Covid have certainly led to an acceleration thanks to the growth of interest in online trading (recorded by a recent ByTek analysis on Google search) driven by the spread of zero-commission platforms (RobinHood in America or Etoro in Italy), the time freed up in lockdown periods and the new financial independence of digital natives and millennials. 

This change, like all innovative processes, has essential implications for those who operate in the market and in this case the main evidence is that traditional data are no longer enough to evaluate and select investments.

Traditional financial data, which for years has been at the center of evaluations by investment institutions of all kinds, now represents only a partial view, a halfway picture of reality. The GameStop case is the evidence of how traditional data (which induced many hedge funds to short the stock) show obvious limits in understanding the market and promptly intercept risks and opportunities (analyzing the discussions on the social network Reddit it was possible to intercept retail positions on the stock already a few weeks before).

The main limits of traditional data are precisely the inability to intercept the opinion of retail investors, the lack of reactivity in a changing world that does not wait for quarterly data and the inability to identify those nuances and weak signals that, if caught, can impact the market (think of Elon Musk’s posts).

 Alternative data contribute to make up for these shortcomings. Alternative data are all those data coming from digital environments, such as search engines, forums, social networks, blogs, consumers reviews, e-commerce platforms, maps and so on. This is mostly unstructured data, i.e. originally text, images, videos, with a much higher update rate than traditional data. The process of extracting and weighing useful information from this data is obviously very complex given the truly big amount, the need to link it to a financial entity and to eliminate the inevitable noise and fake news. It can only be done through the use of Artificial Intelligence techniques (Natural Language Processing, Sentiment Analysis, Entity recognition) capable of converting data quantity into information quality.

The use of alternative data therefore provides a competitive advantage to those operating in the market but it is also necessary to have the ability to translate it into actionable metrics. FinScience, an Italian company part of the Datrix group founded by former Google with the active contribution of shareholders from investment banking, has been working in this field since 2017 and has developed an artificial intelligence-based solution capable of scanning the digital world for alternative data valuable for the investment world. FinScience collects data daily from various media sources (news sites, blogs, forums) but also from the vast world of social networks such as Twitter, Reddit or Stocktwits (social media vertical for the investment world) and transforms them into popularity metrics, sentiment values, and weak signals of risk or opportunity associated with listed securities. 

But even alternative data alone is not enough to make investment decisions. The best results are achieved with a Data Integration approach, which is able to combine the partial representation provided by traditional data with assessments and analysis from the digital world and alternative data. FinScience offers this activity through a platform available to both institutional and private investors. Over the past year, FinScience has published several stock selections based on this integration applied to trends and events with high market impact. 

Examples include the stock selection related to the Green Economy (+294% in about a year), businesses impacted by 5G (about +60%) or the stock selection associated with Biden (+130% from July 2020 to January 2021). All FinScience selections have outperformed the general and sectoral market indices. The selections are updated semi-annually in order to eliminate stocks with declining popularity and sentiment and insert new ones correlated to emerging sub-trends. As in the case of the MedTech selection, which saw the entry of the company Nanthealth, which distinguished itself at the beginning of December with peaks of interest associated with very positive news that in the following days turned into peaks in price (+126% in one day) and volume. 

ALTERNATIVE DATA INTELLIGENCE BY FINSCIENCE

Datrix Data Breakfast

Datrix newsletters about Data and AI applications

Trends, news, case studies and insights on the application of Data and Artificial Intelligence to Marketing & Sales, Investing and Publishing

Subscribe now