Skip to contents

Map survey age composition correlation parameters

do_SrvAge_corr_pars_mapping.Rd

Internal helper called by Setup_Mod_SrvIdx_and_Comps to construct the TMB/RTMB factor maps for SrvAge_corr_pars [n_regions × n_sexes × n_srv_fleets × 2] and SrvAge_corr_pars_agg [n_srv_fleets], the correlation parameters for logistic-normal survey age composition likelihoods.

Usage

do_SrvAge_corr_pars_mapping(input_list)

Arguments

input_list

Named list with $data, $par, and $map sublists. Requires $data$n_srv_fleets, $data$n_regions, $data$n_sexes, $data$SrvAgeComps_Type, $data$SrvAgeComps_LikeType, and $data$UseSrvAgeComps.

Value

The input input_list with $map$SrvAge_corr_pars and $map$SrvAge_corr_pars_agg set to factor vectors. Active parameters receive sequential integer indices; inactive parameters are NA.

Details

The trailing dimension of SrvAge_corr_pars distinguishes the age AR1 correlation (index 1, used by likelihoods 3 and 4) from the sex correlation (index 2, used only by likelihood 4). Parameters are activated conditionally on the fleet's likelihood type and composition type: aggregated compositions (type = 0) activate SrvAge_corr_pars_agg only for likelihood 3; split or joint-by-sex compositions activate SrvAge_corr_pars with index 2 skipped when n_sexes = 1. Fleets using multinomial likelihoods or with no active age composition data have all parameters mapped to NA.

See also

do_SrvAge_theta_mapping for the overdispersion parameter mapping; do_SrvLen_corr_pars_mapping for the analogous length composition correlation mapping.