Data-driven prediction of thresholded time series of rainfall and self-organized criticality models

Together with co-authors with Anna Deluca and Álvaro Corral, Nicholas has studied the predictability of thresholded events in rain data and the Manna model. Specifically, a running hazard function is used as a decision variable in order to predict future events, and the performance of the method is evaluated via a receiver operating characteristic. In the case of the SOC sandpile model, the scaling of quiet-time distributions with increasing threshold leads to increased predictability of extreme events. For rainfall data, the quiet-time distributions do not scale for high thresholds, which means that the corresponding ROC curves cannot be straightforwardly related to those for lower thresholds. In this way, ROC curves are useful for highlighting differences in predictability of extreme events between toy models and real-world phenomena.
The article is published in Physical Review E:
http://journals.aps.org/pre/abstract/10.1103/PhysRevE.91.052808