The Stats-Machine Learning-Econometrics Seminar.
Time: 10:30 am – 11:30 am
Date: 21st of May 2019
Place: Room 3001.
Jérémy Fouliard (LBS) – “Answering the Queen : online machine learning and financial crises”, joint work with Hélène Rey (LBS)

Abstract: Predicting systemic financial crises has been elusive. It is however a key input for macroprudential policy. We consider the framework of sequential predictions, also called  online machine learning,   to predict systemic financial crises in a panel of 7 countries. Since there is no consensus on the origins of financial crises, we optimally aggregate a class of reference forecasters using 4 types of aggregation rules. We can predict crisis periods 3 years in advance out-of-sample with remarkable accuracy. We report for each country the subset of variables that appear the most frequently in our best forecasting models.