Build a csfmt_ensemble_v4

Builds the ensemble that the v4 methods take and return. It holds one row per series and week in $data, and one draw matrix per measure in $draws.

Usage

csfmt_ensemble_v4(data, id_cols, time_col = "isoyearweek", draws = list())

Arguments

data

A data.table with the identity columns and time_col.

id_cols

The identity columns that define one series.

time_col

The column that orders time within a series.

draws

A named list of matrices, one per measure, with nrow(data) rows in the row order of data.

Value

A csfmt_ensemble_v4: a list with data, draws, id_cols, time_col and state. The state is "full", which means that $draws holds the draws.

Details

The constructor copies data and adds time_series_id and time_series_label. It sorts the rows by series and time_col, adds time_series_internal_id, 1..n within each series, and puts the draw rows in the same order. The names of the columns and of the draw matrices MUST be at most 63 bytes, and MUST differ in their first 63 bytes.

See also

Other v4 functions: as_ensemble_v4(), csfmt_reporting_triangle_v4(), ens_add_rate_v4(), ens_compress_v4(), mem_thresholds_v4(), nowcast_ecdf_v4(), nowcast_passthrough_v4(), print.csfmt_ensemble_v4(), short_term_trend_v4(), signal_detection_hlm_v4()

Examples

d <- data.table::data.table(
  location_code = "nation",
  age = "total",
  isoyearweek = c("2023-01", "2023-02", "2023-03")
)
set.seed(1)
ens <- csfmt_ensemble_v4(
  d,
  id_cols = c("location_code", "age"),
  draws = list(numerator = matrix(rpois(3 * 100, 20), nrow = 3))
)
ens
#> <csfmt_ensemble_v4> full | 3 rows | 1 series | draws: numerator
ens$state
#> [1] "full"