Flatline forecaster argument constructor — flatline_args

Constructs a list of arguments for flatline_forecaster().

Usage

flatline_args_list(
  ahead = 7L,
  n_training = Inf,
  forecast_date = NULL,
  target_date = NULL,
  quantile_levels = c(0.05, 0.95),
  symmetrize = TRUE,
  nonneg = TRUE,
  quantile_by_key = character(0L),
  nafill_buffer = Inf,
  ...
)

Arguments

ahead: Integer. Unlike arx_forecaster(), this doesn't have any effect on the predicted values. Predictions are always the most recent observation. However, this does impact the residuals stored in the object. Residuals are calculated based on this number to mimic how badly you would have done. So for example, ahead = 7 will create residuals by comparing values 7 days apart.
n_training: Integer. An upper limit for the number of rows per key that are used for training (in the time unit of the epi_df).
forecast_date: Date. The date on which the forecast is created. The default NULL will attempt to determine this automatically.
target_date: Date. The date for which the forecast is intended. The default NULL will attempt to determine this automatically.
quantile_levels: Vector or NULL. A vector of probabilities to produce prediction intervals. These are created by computing the quantiles of training residuals. A NULL value will result in point forecasts only.
symmetrize: Logical. The default TRUE calculates symmetric prediction intervals. This argument only applies when residual quantiles are used. It is not applicable with trainer = quantile_reg(), for example.
nonneg: Logical. The default TRUE enforces nonnegative predictions by hard-thresholding at 0.
quantile_by_key: Character vector. Groups residuals by listed keys before calculating residual quantiles. See the by_key argument to layer_residual_quantiles() for more information. The default, character(0) performs no grouping. This argument only applies when residual quantiles are used. It is not applicable with trainer = quantile_reg(), for example.
nafill_buffer: At predict time, recent values of the training data are used to create a forecast. However, these can be NA due to, e.g., data latency issues. By default, any missing values will get filled with less recent data. Setting this value to NULL will result in 1 extra recent row (beyond those required for lag creation) to be used. Note that we require at least min(lags) rows of recent data per geo_value to create a prediction. For this reason, setting nafill_buffer < min(lags) will be treated as additional allowed recent data rather than the total amount of recent data to examine.
...: Space to handle future expansions (unused).

Value

A list containing updated parameter choices with class flatline_alist.

Examples

flatline_args_list()
#> • ahead : 7
#> • n_training : Inf
#> • forecast_date : "NULL"
#> • target_date : "NULL"
#> • quantile_levels : 0.05 and 0.95
#> • symmetrize : TRUE
#> • nonneg : TRUE
#> • quantile_by_key : "_empty_"
#> • nafill_buffer : Inf
flatline_args_list(symmetrize = FALSE)
#> • ahead : 7
#> • n_training : Inf
#> • forecast_date : "NULL"
#> • target_date : "NULL"
#> • quantile_levels : 0.05 and 0.95
#> • symmetrize : FALSE
#> • nonneg : TRUE
#> • quantile_by_key : "_empty_"
#> • nafill_buffer : Inf
flatline_args_list(quantile_levels = c(.1, .3, .7, .9), n_training = 120)
#> • ahead : 7
#> • n_training : 120
#> • forecast_date : "NULL"
#> • target_date : "NULL"
#> • quantile_levels : 0.1, 0.3, 0.7, and 0.9
#> • symmetrize : TRUE
#> • nonneg : TRUE
#> • quantile_by_key : "_empty_"
#> • nafill_buffer : Inf