Skip to contents

Estimate Rt with a smooth derivative

Source: R/model_infections.R
R_estimate_smooth_derivative.Rd

This option estimates the effective reproduction number Rt over time as a smooth trend. It uses sparse smoothing splines placed on the first-order differences of Rt.

Usage

R_estimate_smooth_derivative(
  R_start_prior_mu = 1,
  R_start_prior_sigma = 0.8,
  spline_knot_distance = 14,
  trend_prior_shape = 5,
  trend_prior_scale = 0.005,
  link = "inv_softplus",
  R_max = 6,
  modeldata = modeldata_init()
)

Arguments

R_start_prior_mu

Prior (mean) on the initial reproduction number (intercept).

R_start_prior_sigma

Prior (standard deviation) on the initial reproduction number (intercept).

spline_knot_distance

Distance (in days) between spline knots for the penalized smoothing splines. Shorter distances increase flexibility of the Rt trajectory but also the daily Rt uncertainty.

trend_prior_shape

Normal-Exponential-Gamma (NEG) prior (shape) for the Rt trend. The NEG prior is sparse, i.e. it has a strong peak at zero (transmission remains unchanged) and long tails (allows for occasional large positive or negative changes). Smaller shape parameters will lead to more sparseness (i.e. fewer but sharper changes in Rt). Note that when adjusting the shape, you will likely also have to adjust the scale.

trend_prior_scale

Normal-Exponential-Gamma (NEG) prior (scale) for the strength of the Rt trend. See sharp_changes_prior_shape above for an explanation. Larger scales will lead to more Rt variability.

Link function. Currently supported are inv_softplus (default) and scaled_logit. Both of these links are configured to behave approximately like the identity function around R=1, but become increasingly non-linear below (and in the case of scaled_logit also above) R=1.

R_max

If link=scaled_logit is used, a maximum reproduction number must be assumed. This should be higher than any realistic R value for the modeled pathogen. Default is 6.

modeldata

A modeldata object to which the above model specifications should be added. Default is an empty model given by modeldata_init(). Can also be an already partly specified model returned by other EpiSewer modeling functions.

Value

A modeldata object containing data and specifications of the model to be fitted. Can be passed on to other EpiSewer modeling functions to add further data and model specifications.

The modeldata object also includes information about parameter initialization (.init), meta data (.metainfo), and checks to be performed before model fitting (.checks).

Details

The priors of this component have the following functional form:

  • R_start (intercept): Normal

  • trend_prior: Normal-Exponential-Gamma

See also

Other Rt models: R_estimate_approx(), R_estimate_changepoint_splines(), R_estimate_ets(), R_estimate_gp(), R_estimate_piecewise(), R_estimate_rw(), R_estimate_splines()