About
Mortality Watch aggregates mortality data from multiple sources to generate daily updated mortality charts. Select from many jurisdiction's and different chart types. The underlying dataset is now the largest and most detailed on the web. Data and calculations (R language) are open source available on GitHub. Crude mortality rate (CMR) is deaths per one hundred-thousand residents. Age Standardized Mortality Rate (ASMR) is calculated using the WHO2015, ESP2013, US2000 Standard Population and each jurisdictions 2020 age table.
Caution: Excess Deaths are based on the baseline model, which may or may not reflect reality.
Excess Model Baseline
For each jurisdiction, period of time and metric, pre-pandemic data is back tested to choose the baseline length n with the lowest root mean squared error (RMSE) for a four to ten-year period of a four-year forecast. A linear regression model is used (fable::TSLM + trend()) with a seasonal parameter added for sub-year resolutions.
Additionally to the default model, users can choose from the following baseline methods:
- Last Value: fable::NAIVE
- Average: fable::TSLM
- Linear Regression: fable::TSLM + trend()
- Last Value: fable::ETS + error() + trend()
A seasonal parameter is added for sub-year resolutions.
Contact
Feedback or questions via: E-mail or direct message via @MortalityWatch on X.
More
More information can be found here: More Charts · Data · Open Source