Pierre-Philippe Combes, Gilles Duranton, Laurent Gobillon, Clément Gorin, Yanos Zylberberg, 17 November 2020

Applying machine learning to rich historical data sources provides the opportunity to draw novel insights for fields such as urban and spatial economics. Using evidence from France, this column shows how such information might be derived from historical maps to shed new light on the growth of towns and agglomerations, and could inform our understanding of various human behaviours from community evolution to agricultural productivity.


CEPR Policy Research