EpiNow2 is an R package for estimating the time-varying reproduction number, and growth rates based on underlying latent infections. It is also used to forecast the future course of outbreaks both in terms of forecasting the reproduction number and reported cases. Community support is needed to allow the developers to focus on developing, validating and improving the underlying methodology.
EpiNow2 fits a latent state gaussian process model by default that can produce fitting issues, especially when using challenging real-time data. Users would benefit from automated extraction of fit summary statistics passed into a format that is easy to digest. Users running at scale (including the developers) would benefit from this runtime information being summarised across subregions (i.e when estimating regional Rt for a country).
Tasks:
Add summarised runtime information to estimate_infections
Add a function to summarise runtime information across epinow/estimate_infection calls
Add a call to this function to regional_epinow
[impact: please copy in from previous]
[originally proposed by @seabbs]
[suggested repo: https://github.com/epiforecasts/EpiNow2 | @seabbs]
[additional notes: ]
EpiNow2 is an R package for estimating the time-varying reproduction number, and growth rates based on underlying latent infections. It is also used to forecast the future course of outbreaks both in terms of forecasting the reproduction number and reported cases. Community support is needed to allow the developers to focus on developing, validating and improving the underlying methodology.
EpiNow2 fits a latent state gaussian process model by default that can produce fitting issues, especially when using challenging real-time data. Users would benefit from automated extraction of fit summary statistics passed into a format that is easy to digest. Users running at scale (including the developers) would benefit from this runtime information being summarised across subregions (i.e when estimating regional Rt for a country).
Tasks:
[impact: please copy in from previous] [originally proposed by @seabbs] [suggested repo: https://github.com/epiforecasts/EpiNow2 | @seabbs] [additional notes: ]