EpiEstim model for fable — EPIESTIM • acciddasuite

A fable model that estimates the reproduction number Rt with EpiEstim::estimate_R() and projects future incidence with projections::project(), for use inside model. The aggregation period is read from the tsibble index; Rt is estimated on the most recent data and used to simulate n_sim forward paths, returned as sample distributions.

Usage

EPIESTIM(
  formula,
  mean_si,
  std_si,
  rt_window = 14L,
  n_sim = 100L,
  R_fix_within = TRUE
)

Arguments

formula: Response variable, e.g. observation. Exogenous regressors are not supported.
mean_si, std_si: Mean and SD of the serial interval, in days.
rt_window: Sliding window width (days) for Rt estimation. Smaller tracks recent trends; larger is smoother. Default 14.
n_sim: Number of simulated forecast paths. Default 100.
R_fix_within: If TRUE, hold Rt constant within each simulated path (recommended for short horizons).

Value

A model definition for use inside model.