Empirical Models to Predict Rare High-Impact Events

For several decades scientists from different fields have realized that many features of the natural and human world do not follow Gaussian distributions, ie, they don’t cluster neatly around a mean. On the contrary, quantities such as the magnitude of Earthquakes, the income of individuals, the number of Facebook friends or the word frequency have “heavy tail” distributions. That means that while there are many instances of weak Earthquakes and many poor people, from time to time there are a few extremely devastating Earthquakes and some billionaires.

Very often, underlying these processes, we find exponential growth (the rich get richer) and multiplicative noise. We are developing general modelling approaches that allow estimating parameters of the processes underlying heavy tails. This will allow us to 1) Gain insights into the origin of heavy tails; 2) use powerful methods such as generalized linear models to look for explanatory factors and 3) devise strategies to reduce disparities where they are not desirable.

Understanding of such dynamics is relevant to study community formation and can be applied to the study of disinformation or how ideas take hold.

This project is currently funded by an ERC Starting Grant.

Starting Date: January 2019