This function takes in the combined set of wastewater and hospital admissions quantiles by forecast date, date, and location, creates a dataframe formatted as it would have been for submission to the COVID forecast Hub, and saves this to disk if specified. It returns the metadata on the number of locations total and the number of locations using the wastewater model for each forecast date submission saved. It assumes the wastewater-informed model output quantiles passed in have already been filtered to remove ones that we would wish to exclude (for convergence diagnostics or other reasons)
Usage
create_hub_submissions(
hosp_quantiles_ww,
hosp_quantiles_hosp,
forecast_dates,
hub_subdir,
model_name,
scenario = "status_quo",
save_files = TRUE
)Arguments
- hosp_quantiles_ww
a dataframe of the probabilistic estimates of hospital admissions formatted as quantiles, from the wastewater model, for multiple forecast dates if present for a particular location
- hosp_quantiles_hosp
a dataframe of the probabilistic estimates of hospital admissions formatted as quantiles, from the hospital admissions model, for multiple forecast dates
- forecast_dates
vector of forecast dates that we want to create mock submissions for
- hub_subdir
character string of the outer directory specifying where to write the submission files to (which will be named by forecast date)
- model_name
character string indicating the name of the model. This will be used to determine both the name of the submission
.csvfile and the name of its enclosing directory, per COVID-19 Forecast Hub formatting.- scenario
string indicating which wastewater availability scenario to use when creating the hub submission file. Default
"status_quo"(all actually available wastewater data).- save_files
Save the created submission data frame to disk as a
.csvfile? Boolean, defaultTRUE.