Evaluates the performance of the nowcast function on historic data.
Arguments
- nowcast_object
Object from the function nowcast
- n_week_adjusting
Number of weeks to adjust
Value
Residualplots for all ncor_i and some evaluationmetrixs for each of them as well as a plot containing credible intervals using the simulations
Details
For more details see the help vignette:
vignette("nowcast", package="attrib")
Examples
data_aggregated <- data.table::as.data.table(data_fake_nowcasting_county_aggregated)
data_aggregated <- data_aggregated[location_code == "county_nor03",]
n_week_training <- 50
n_week_adjusting <- 5
date_0 <- data_aggregated[nrow(data_aggregated),]$cut_doe
nowcast_object <- nowcast(data_aggregated,offset = "log(pop)",
n_week_adjusting,n_week_training,date_0,
nowcast_correction_fn = nowcast_correction_fn_quasipoisson,
nowcast_correction_sim_fn = nowcast_correction_sim_quasipoisson)
nowcast_eval_object <- nowcast_eval(nowcast_object, n_week_adjusting)