An in-development R package and a Bayesian hierarchical model jointly fitting multiple "local" wastewater data streams and "global" case count data to produce nowcasts and forecasts of both observations
This handles only the model file and get_stan_data.R/initialization.R, but downstream things should be minimally affected. In most cases, things should be backward compatible.
Changes:
The reference subpopulation is always the first
Most transformed parameters do not distinguish the reference subpopulation from the others. This is particularly useful since it makes it easier to move to a situation in which, e.g., more than one subpopulation has no wastewater.
Offsets now have more verbose names and aren't named $m$ (which could cause confusion with slopes, whereas they're more like intercepts
This handles only the model file and
get_stan_data.R/initialization.R, but downstream things should be minimally affected. In most cases, things should be backward compatible.Changes: