Predictions based on DRM.

Considering a new dataset (across the same sites), computes predictions based on the DRM passed as drm.

Usage

# S3 method for class 'adrm'
predict(
  object,
  new_data,
  past_data,
  f_test,
  type = "predictive",
  seed = 1,
  cores = 1,
  ...
)

predict_drm(...)

Arguments

object

An adrm object containing the output from the fit_drm() function.

new_data

a data.frame with the dataset at which we wish to obtain predictions.

past_data

a data.frame with the dataset from the last year used in model fitting. Only needed when f_test is not missing or when estimating survival.

f_test

a matrix informing the instantaneous fishing mortality rates at each age (columns) and timepoint (rows).

type

type of predictions to be computed. Admitted values are

• "predictive" (default): posterior predictive distribution;

• "expected": theoretical mean of the posterior predictive distribution;

• "latent": latent density (i.e., disregarding the observation error);

seed

a seed used for the predictions. predictions are obtained through Monte Carlo samples from the posterior predictive distribution. Therefore, a seed is needed to ensure the results' reproducibility.

cores

number of threads used to compute the predictions. If four chains were used in the drm, then four (or less) threads are recommended.

...

params to be passed to predict.adrm

Value

an object of class "CmdStanGQ" containing samples for the posterior predictive distribution for predictions.

Details

The current version of the code assumes the data where predictions are needed is ordered by "site" and "site" and, in addition, its sites MUST be the same as the ones used to obtain the parameters' estimates from the the drm object.

See also

fit_drm()

Author

lcgodoy