COVID-19: Causal analysis of confinement policies

MLG-ULB participation to the CodeVsCovid19 hackaton

Try it online!

Summary

Identify the impact and effectiveness of government policy on the number of infected and deaths.

Description

We aim to use historical time series data of the number of cases on different countries/regions to leverage insights into the causal impact of the policies applied by the governments. This analysis is applied on a country-by-country basis, and consists in

• Comparing the growth rate in the number of cases before and after a restrictive policy is applied
• Determining when a significant change in the growth rate is measured
• Correlate (possibly with a delay) policy enactment with growth rate change
• Rank countries according to the effectiveness of their policies

The Team

• Théo Verhelst ([email protected]): Team lead
• Jacopo ([email protected]): Data preprocessing and analysis
• Gian Marco Paldino ([email protected]): Shiny Dashboard
• Elias Fernández ([email protected]): Data preprocessing & literature research
• Mattia Bontempi: Data collection
• Gianluca Bontempi ([email protected]): Mentor

The Code

The code of this dashboard is open-source and available on GitHub.

Data Sources

• Global time series: JHU CSSE
• Italy time series: Presidenza del Consiglio dei Ministri - Dipartimento della Protezione Civile
• Country policy dates: a Kaggle dataset

Built With

• jupyter-notebook
• r
• shiny