Skip to contents

Map fishery composition correlation parameters

do_comp_corr_pars_mapping.Rd

Constructs factor maps for a composition correlation-parameter array (e.g. FishAge_corr_pars, region- and sex-specific AR1/sex correlation) and its aggregated counterpart (e.g. FishAge_corr_pars_agg) for 1D and 2D logistic-normal composition likelihoods. Parameters are activated only when the corresponding LikeType is in c(3, 4, 6, 7) (1D and 2D logistic-normal, and the same two with the zeros dropped); all other likelihoods, or fleets with no observed compositions, map correlation parameters to NA. For the 2D logistic-normal (LikeType == 4 or 7), both trailing elements of the [...,2] slice are activated: element 1 for the AR1 coefficient and element 2 for the sex correlation (skipped when n_sexes == 1).

Usage

do_comp_corr_pars_mapping(
  input_list,
  comp_prefix,
  discard = FALSE,
  has_pop = FALSE,
  fleet_field = "n_fish_fleets"
)

Arguments

input_list

Named list with $data, $par, and $map sublists.

comp_prefix

Character, either "FishAge" or "FishLen".

discard

Logical. If TRUE, maps the discard variant. Default FALSE.

has_pop

Logical. If TRUE, maps the population-specific variant. Default FALSE.

fleet_field

Character. Name of the $data field giving the number of fleets to loop over. Default "n_fish_fleets"; pass "n_srv_fleets" for survey composition types ("SrvAge", "SrvLen"), which have no discard variant.

Value

The input input_list with the corresponding $map$<...>_corr_pars and $map$<...>_corr_pars_agg set to factor vectors.

Details

One function serves every composition block. comp_prefix selects age versus length and fishery versus survey, discard selects the retained or discarded data source, has_pop selects the aggregated or population specific data source, and fleet_field names the fleet count to size the map by. Same parameterization as do_comp_theta_mapping.