An evolving set of Bayesian Hierarchical models being used to model the spread of COVID-19 and its potentially dependent factors. There is an accompanying collection of scripts to collect covid19 and other related data.
An Arxiv pre-pint was submitted on 10 July 2020. The code used to do the analysis of this paper resides in this repository. Currently the repository is a bit messy so a brief explanation is given below to help readers navigate to the important sections.
All notebooks in this repository reside in the notebooks directory. The main notebooks of interest to readers of the paper begin with 4.1.1.xxxx. The notebooks generally have mostly explanatory names for which model is contained inside. For complete clarity the notebook definitions are given below.
| Model | Notebook |
|---|---|
| Time varying growth | 4.1.1.10 Sigmoid Growth Regression.ipynb |
| No Factors | 4.1.1.3 No Params Regression.ipynb |
| BCG vaccine coverage | 4.1.1.1 BCG Vaccine Coverage Regression.ipynb |
| Temperature | 4.1.1.2 Temperature Regression.ipynb |
| Relative Humidity | 4.1.1.3 Humidity Regression.ipynb |
| UV Index | 4.1.1.3 UVindex Regression.ipynb |
| Tests per 1000 (testing) | 4.1.1.4 Test Rate Regression.ipynb |
| Positive Rate (testing) | 4.1.1.5 Positive Test Rate Regression.ipynb |
| All excluding testing | 4.1.1.12 BCG and Climate(incl. UV).ipynb |
| All including testing | 4.1.1.11 BCG, Climate(incl. UV) and Testing.ipynb |
| Model | Notebook |
|---|---|
| Blood | 4.1.1.7 A+ Blood Type Regression.ipynb |
| All including testing and blood | 4.1.1.11 BCG, Climate(incl. UV), Testing and Blood Regression.ipynb |