Join hubverse forecasts to observed data

Expects forecast output in hubverse format (e.g. as created by get_hubverse_quantile_table()) and an observed data table with location, date, and value columns. The names of these columns the observed data table can be configured; defaults are "location", "date", and "value", respectively (direct correspondence with standard Hub "target data" tables).

Usage

hubverse_table_with_obs(
  hubverse_forecast_table,
  observation_table,
  obs_value_column = "value",
  obs_date_column = "date",
  obs_value_name = "observed",
  id_cols = c("location", "target"),
  join = "full"
)

Arguments

hubverse_forecast_table

Forecasts, as a hubverse-format tibble, for instance as produced by get_hubverse_quantile_table(), with columns including target_end_date, and location.

observation_table

observations, as a tibble.

obs_value_column

Name of the column containing observed values in the observed table, as a string. Default "value"

obs_date_column

Name of the column containing date values in the observed table, as a string. Will be joined to the target_end_date column in the hubverse table. Default "date"

obs_value_name

Name for the column of observed values in the resulting table (since "value" clashes with the forecast value column in a standard hubverse table. Default "observed".

id_cols

Additional ID columns to join on, in addition to obs_date_column. Default c("location", "target").

join

Which SQL-style dplyr mutating join function to use when joining the tables. Options are "full" (for dplyr::full_join()), "left" (dplyr::left_join()), "right" (dplyr::right_join()), and "inner" (dplyr::inner_join()). In the join, the hubverse forecast table is the left table (x) and the observation table is the right table (y). Default "full" (i.e. keep all forecasts and observations, even if some forecasts do not have a corresponding observation or some observations do not have a corresponding forecast).

Value

A tibble with the observed values added.