Bayesian Nowcasting with application to COVID-19 fatalities in Sweden

This repository contains data and code for the article Flexible Bayesian Nowcasting with application to COVID-19 fatalities in Sweden by Fanny Bergström, Felix Günther, Michael Höhle and Tom Britton at Stockholm University. The artice is published in PLOS Comp Bio.

Monthly updated nowcast estimates of COVID-19 fatalities and ICU admissions in Sweden using our proposed method are found at

https://staff.math.su.se/fanny.bergstrom/covid19-nowcasting.

Repo structure

The implementation of the different Nowcasting models are given in the folder stan_models.

The R code used for our analysis is structures as follows

1. Data preprocessing.
2. Definition of functions for fitting the nowcast models to given reporting dates for each model and saving the results of the post-processing of a fit.
3. Estimation of chosen nowcast models and reporting dates.
4. Summarize results of the nowcasts.
5. Plots (incl. results of single days).