Vous êtes ici

A Machine-Learning Approach to Nowcast the GDP in Sub-Saharan Africa

Author(s): 
Karim Barhoumi
Seung Mo Choi
Tara Iyer
Jiakun Li
Franck Ouattara
Andrew Tiffin
Jiaxiong Yao

In order to aid policy decisions during the COVID-19 pandemic, we propose a machine learning approach to nowcast GDP growth in sub-Saharan Africa, a region where official statistics are released with considerable delays. We show that machine learning methods provide nowcasts with lower root mean square errors than standard benchmarks used in the literature. Nowcasts imply that the COVID-19 crisis initially had a large adverse impact on the region.