Checks that a data frame has the required columns and structure for
use with get_ncast and get_fcast.
Returns a typed accidda_data object that downstream functions
can accept without repeating validation.
Arguments
- df
A data frame (or tibble) with at least:
target_end_date (Date), observation (numeric),
location (character), and target (character).
An optional as_of (Date) column enables nowcasting via
get_ncast.
Value
An accidda_data object (a list) with:
- data
The validated data frame with coerced date types.
- location
Single location identifier.
- target
Single target identifier.
- window
Named vector with from and to dates.
- history
Logical. TRUE if revision history
(as_of) is present.
Examples
if (FALSE) { # \dontrun{
# From get_data
df <- get_data("covid", "ny") |> check_data()
# User-provided data
my_df <- read.csv("my_data.csv") |> check_data()
# Then into the pipeline
df |> get_fcast(eval_start_date = "2025-01-01")
} # }
