As part of the Vilfredo Pareto Research Seminar series, the International Economics Department at the Graduate Institute is pleased to invite you to a public talk given by Cong Peng, PhD Researcher at Harvard University.
He will present his work titled Roads to Development? Examining the Zambian Context Using AI.
Abstract: With the backdrop of ambitious road network expansion and a growing fiscal deficit, this study examines the impact of improvements in Zambia’s road network condition over the years 2009 and 2019. We trained a Convolutional Neural Network model on 6000 high-resolution satellite images to predict the pavement status of roads and changes in built-up area across the country. This modelling provides comprehensive tracking of changes in road condition, and therefore predicted travel time and each area’s access to markets, over time. The study finds that improvements in market access drives expansion of urban built-up area. Integrating data from the Demographic and Health Surveys, the study also finds that improvements in market access drives improvements across a range of household socioeconomic outcomes. On the other hand, after integrating additional sources of remote sensing data, the study finds that the increasing urban development comes at a cost of higher levels of air pollution.
About the speaker
Cong Peng is a PhD Researcher at Harvard University. His research features in using real-time data to solve pressing problems in economics, by connecting advances in data science with econometrics. His work revolves around applying state-of-art causal inference methods to draw inference from Big Data, which includes data generated from the Internet of Things (IoT) platforms, user-generated content, open source GIS data, and high-resolution satellite data.