Convert a baselinenowcast_df object to a forecast_point object

This function converts a point baselinenowcast_df object as returned by baselinenowcast() with output_type = "point" to a forecast_point object which can be used for scoring with the scoringutils package.

Usage

# S3 method for class 'baselinenowcast_df'
as_forecast_point(
  data,
  latest_obs,
  observed = "count",
  model = "baselinenowcast",
  ...
)

Arguments

data

A baselinenowcast_df object as returned by baselinenowcast() with output_type = "point".

latest_obs

A data.frame containing the truth to score against, with one row per (reference_date, strata) combination. Must contain a reference_date column and a column with observed counts named according to observed (default "count"). Additional columns shared with data (such as strata columns) are used as merge keys. latest_obs should hold the reported total at each reference date evaluated at the same max_delay horizon used for the nowcast (rolling truth), not the partial total available when the nowcast was run.

observed

Character string giving the name of the column in latest_obs that holds the observed value. Defaults to "count" to match the input format of baselinenowcast.data.frame().

model

Character string used as the value of the model column on the returned forecast object. Lets scoringutils::summarise_scores() run with its default by = "model". Defaults to "baselinenowcast". Pass NULL to omit the column (e.g. when data already carries its own model column).

...

Additional arguments passed to scoringutils::as_forecast_point().

Value

A forecast_point object as returned by scoringutils::as_forecast_point().

Details

The nowcast point estimates in data are merged with the latest observations in latest_obs on reference_date and any other shared columns and then passed to scoringutils::as_forecast_point(), using pred_count as the predicted value and the column named by observed as the observed value.

See also

Main nowcasting interface functions as_forecast_sample.baselinenowcast_df(), assert_baselinenowcast_df(), baselinenowcast(), baselinenowcast.data.frame(), baselinenowcast.reporting_triangle(), baselinenowcast_df-class, new_baselinenowcast_df()

Examples

if (FALSE) { # interactive() && requireNamespace("scoringutils", quietly = TRUE)
library(scoringutils)

max_delay <- 25
nowcast_date <- as.Date("2026-04-01")

full_tri <- as_reporting_triangle(syn_nssp_df) |>
  truncate_to_delay(max_delay = max_delay)

n_drop <- sum(as.Date(rownames(full_tri)) > nowcast_date)
rep_tri <- full_tri |>
  truncate_to_row(t = n_drop) |>
  apply_reporting_structure() |>
  tail(n = 40)

nowcast <- baselinenowcast(rep_tri, output_type = "point")

# Rolling truth from the same full triangle. `as_forecast_point()` scores
# only the right-truncated nowcast dates (the `nowcast` column); the
# inner-join merge restricts the truth to those dates.
truth_df <- as.data.frame(full_tri)
latest_obs <- aggregate(count ~ reference_date, data = truth_df, FUN = sum)

fp <- as_forecast_point(nowcast, latest_obs)
fp
scores <- score(fp)
scores
summarise_scores(scores)
}