CPUE index likelihood (multi-index) — get_cpue

Computes the likelihood for one or more standardised CPUE/survey indices using a log-linear model. Each index has its own catchability (q), extra variance (tau), power parameter (omega), and effort creep. Mean-centering of predicted CPUE is performed within each index. Time steps are one-based: (year_index - 1) * n_season + season_index.

Usage

get_cpue_like(
  cpue_data,
  parameters,
  number_ysa,
  sel_fya,
  weight_fya,
  cpue_switch = 1L
)

Arguments

cpue_data

a list of data inputs. Must contain:

cpue_data: data.frame with columns ts, fishery, value, se, units, and index.
cpue_switch: integer switch (0 = skip likelihood).
n_index: integer number of distinct indices.

parameters

a list of parameter values. Must contain:

log_cpue_q: numeric vector [n_index].
log_cpue_tau: numeric vector [n_index].
log_cpue_omega: numeric vector [n_index].

number_ysa

a 3D array [n_year, n_season, n_age] of numbers-at-age.

sel_fya

a 3D array [n_fishery, n_year, n_age] of selectivity by fishery, year, and age.

weight_fya

a 3D array [n_fishery, n_year, n_age] of weight-at-age by fishery and year.

cpue_switch

boolean flag to calculate the cpue likelihood.

Value

numeric vector of length nrow(cpue_data) with per-observation negative log-likelihood contributions.