Building a coronaradar
summary
A working coronaradar that pulls EU CDC spreadsheets into pandas and reproduces country curves, double-log linearization of cumulative cases to fit a Gumbel-type curve, and iterative decomposition with earth (MARS) segmentation to detect multiple waves. It smooths weekly reporting artifacts, projects one week ahead, and renders an animated folium heatmap of regional forecasts. Example output assigns about 220000 cases to the UK first wave. It generalizes well in practice but remains sensitive to reporting glitches, changes in testing and policy, so projections are conditional not definitive.
AI-written summary · about
chapters (11)
- 0:00CoronaRadar project introduction
- 1:08Project overview and resources
- 8:48Python data science journey
- 14:00Data prep and visualization basics
- 22:05Outbreak curves and flattening
- 30:08Curve fitting with Gumbel
- 41:05Testing increases and data spikes
- 47:00Multiple outbreak waves analysis
- 53:13Netherlands local data analysis
- 57:06Animated CoronaRadar map
- 1:00:57Conclusion and next steps
