Video/Audio Open Access

Novel Application in Machine Learning: Predicting the Issuance of COVID-19 Stay-at-Home Orders in Africa

Rhea, Carter; Mansell, Jordan; Murray, Gregg

In the work, we investigate which factors are most important for the issuance of SAHO among n=54 African countries between January 31st and June 15th, 2020. We employ a novel dataset of 260 different variables capturing country-level information on economic, political, social, external, and health-related factors for each country in our dataset. To identify the most significant factors, we treat the question of which countries issue SAHO as a classification problem (issued orders vs. did not issue orders) and utilize a random forest classifier, a method of machine learning, to identify the variables that best explain whether a country issued a SAHO order or not.

Files (12.8 MB)
Name Size
12.8 MB Download
All versions This version
Views 3737
Downloads 33
Data volume 38.4 MB38.4 MB
Unique views 3434
Unique downloads 33


Cite as