Forecast Revenue and earnings outcome Using Event-Driven Score

One major use of an event’s direction using NLP and big-data is the prediction of quarterly financial performance which reflects the company’s operational activities. In this case study, we tested the 2Q 2018 financial announcements using an event-driven score aggregation to predict companies’ 2Q financial performance. To do so, we generated an event-driven score for companies using the period from 04.01-06.30.18 (2Q) as well as 01.01-06.01.18 (H1). The scoring will contain an average of all the events (each event has its own scoring which reflects the potential impact of the event on the company’s operation) that took place during these periods.

We aim to test the predictive power of a 6-month (H1) event-driven score VS a 3-month (2Q) event-driven score of companies which presents revenue or earnings growth in Q2 2018. We took 386 companies which reported positive financial outcomes for the period APR-JUN 2018 and evaluate the 01.01-30.06.18 event-driven score VS the 04.01-06-30.18 event-driven score to see if the 6-month period predict more accurate if these companies will have a growth in their revenue or earnings VS the 3-month period.

Among these companies that present revenue or earnings growth, we found events such as increase in sales, new orders, new products releases, new JV and M&A activities, increase in demand, new pricing models, entering new markets, and positive ESG events during the past quarter.

The following table contains a list of companies that reported revenue and or earnings growth for the APR-JUN period. We use 2 types of event-driven scores, first is the 6-month period using an aggregation of all events during the 01.01-06.30.18 VS the event-driven score for the period 04.01-06.30.18 period. When using a longer period, the model uses more events to calculate the event-driven score, we assume that more events will lead to higher accuracy.

It is critical to distinguish between event’s pricing decay and operational decay. The impact of events on the share price is decay over time as the current price should reflect all the data which is available, however, the impact on the company operational activity remains for a longer period which will impact the company’s financial performance. Therefore, adding more events into the model should provide more accuracy to predict the company’s operational activities which will be reflected in their financial outcomes.

[table id=13 /]

From the 386 companies that presented a positive financial outcome (presents revenue or earnings growth) for the 2Q 2018, 317 (82%) companies received an event-driven score above 0.52 using the 6-months period VS 305 (79%) using the 3-months period which indicates a bullish trend (the event-driven score is scaled between 0-1, where below 0.48 points to a bearish event-driven score based on events during the filtered period, and above 0.52 points to a bullish event-driven score).

The data shows that longer selected periods for generating an event-driven score impact the accuracy of the model. This can be explained by adding more events to the model which reflects the company’s operating activities.

To conclude, using events to predict a company’s performance, provide additional prediction power on top of other analytic processes when evaluating new positions.

Using big data and NLP technologies to capture alpha by collecting, structuring, and revealing events from news articles, press releases, and financial social media.

(Views and recommendations given in this section are for research purposes only. Please consult your financial adviser before taking any position in the stock/s or currencies mentioned.) Neither First to invest. nor any of its officers, employees, representatives, agents or independent contractors are, in such capacities, licensed financial advisors, registered investment advisers or registered broker-dealers. First to invest does not provide investment or financial advice or make investment recommendations. Nothing contained in this communication constitutes a solicitation, recommendation, promotion, endorsement or offer by First to invest of any particular security, transaction or investment.)