Hi ,
"No plan survives first contact with the enemy." - Helmuth von Moltke
Every data scientist is taught, from day 1, to be cautious about
making future predictions based on the past.
I've been in university lectures where the lecturer has literally made the class recite the mantra: "The past is not always a good indicator of the future."
Yet, most of the time, the past is a *good enough* predictor of
the future to make the predictions we want to make.
That is, until something happens and the trajectory of the world changes completely.
However, as with art, "everything is life is also a remix", and there are few events the world has never previously
encountered.
The COVID-19 pandemic may be the first global pandemic you or I have ever experienced, for example, but it was certainly nothing original.
Events themselves have patterns. You just need to look for them. And fortunately, identifying patterns is what
machine learning does best.
This is what event-driven analysis is all about.
I recently had the opportunity to speak to LevelFields CEO Andrew Einhorn who has built a career on applying event-driven analysis to a variety of use cases - most recently to predicting
movements in the US stock exchange.
You can check out our entire conversation at the link below: Episode 45: AI-Powered Investment Insights
Talk again soon,
Dr Genevieve Hayes.
