search

ImperialCollegeLondon / covid19model

Code for modelling estimated deaths and cases for COVID19.
MIT License
944 stars 271 forks source link

Possible incorrect mixing of percentage and unit ratios in mobility forecast #114

Closed payoto closed 4 years ago

payoto commented 4 years ago

Describe the bug It seems to me that the simulate-regional.r incorrectly mixes percentage and unit ratios to simulate the change in mobility. I've included the two lines from the code base below:

https://github.com/ImperialCollegeLondon/covid19model/blob/c1c86a1282bef9dc11de78f6cbb209f5c14f8f4b/Italy/code/utils/simulate-regional.r#L71

https://github.com/ImperialCollegeLondon/covid19model/blob/c1c86a1282bef9dc11de78f6cbb209f5c14f8f4b/Italy/code/utils/simulate-regional.r#L78

The value of 0.2 leads to a projected mobility with value 0.998 of the last 7 days for the forecast period, which is not what is documented (20% increase).