Modelling extensions for multi-location studies in environmental epidemiology

Fully reproducible code and data for the illustrative application in the paper:

Masselot P. and Gasparrini A. (2005) A new modelling framework for multi-location studies in environmental epidemiology. Statistical Methods in Medical Research DOI:10.1177/09622802241313284.

All components of the dataset are provided in the data subdirectory. The subdirectory also contains the script 00_DownloadData.RData that shows how the data were obtained and pre-processed. The only exception is data/tmean.csv.gz, that was extracted using Python from ERA5Land. ⚠️ This script should not be executed, it is here mainly to document the data gathering process.

Execution of the code should start from 11_Packages.R.

1. Scripts 11_Packages.R to 13_PrepData.R setup the modelling and load data.

2. Scripts 21_LocationAgeSpecific.R to 28_Uncertainty.R implement each corresponding package (sub-section) in section 3 of the paper. Each one rely on previous ones, which means they should be ran in order.

3. Scripts A1_DataDesc.R to A4_AlternativePC.R provides additional results corresponding to the supplementary materials.

Finally, the figures subdirectory includes all Figures and Tables of the paper.