The Challenge of Using Epidemiological Case Count Data: The Example of Confirmed COVID-19 Cases and the Weather
Github Repository for Cohen et al study on "The Challenge of Using Epidemiological Case Count Data: The Example of Confirmed COVID-19 Cases and the Weather"
Research by Francois Cohen, Moritz Schwarz, Sihan Li, Yangsiyu Lu and Anant Jani
This version: May 10th, 2020; first version: April 1st, 2020; data update: April 30th, 2020.
The publicly available data on COVID-19 cases provides an opportunity to better understand this new disease. However, strong attention needs to be paid to the limitations of the data to avoid making inaccurate conclusions. This article, which focuses on the relationship between the weather and COVID-19, raises the concern that the same factors influencing the spread of the disease might also affect the number of tests performed and who gets tested. For example, weather conditions impact the prevalence of respiratory diseases with symptoms similar to COVID-19, and this will likely influence the number of tests performed. This general limitation could severely undermine any similar analysis using existing COVID-19 data or similar epidemiological data, which could, therefore, mislead decision-makers on questions of great policy relevance.
- Download the estimation data here.
- Ensure that the data file is called
ESTIMATION_30042020.dta
and lies in the folder Data/raw. - Open and run the code file
Replicate Results.do
. This should create all tables and figures used in our manuscript.
Should you have any general questions or comments, please contact the corresponding author Dr Francois Cohen.
For any technical questions regarding this GitHub Repository, please contact Moritz Schwarz