afscOSA

One-step ahead (OSA) residual diagnostic plots for observed and predicted age and length compositions with an assumed multinomial distribution from assessment models. OSA residuals are computed using the compResidual R library (Trijoulet and Nielsen, 2022).

Installation

Install compResidual and afscOSA using the instructions below.

# install.packages("devtools")

# see https://github.com/fishfollower/compResidual#composition-residuals for
# installation instructions

# TMB:::install.contrib("https://github.com/vtrijoulet/OSA_multivariate_dists/archive/main.zip")
# devtools::install_github("fishfollower/compResidual/compResidual")

devtools::install_github("noaa-afsc/afscOSA", dependencies = TRUE)

Examples

There are code examples for GOA pollock (a bespoke ADMB model), BSAI Atka mackerel (AMAK), and AI Pacific cod (SS3).

Example scripts are downloaded when afscOSC is installed. Locate them on your computer by running the following commands:

(afscOSA_path <- find.package('afscOSA'))
(afscOSA_examples <- file.path(afscOSA_path, 'examples'))
list.files(afscOSA_examples)

# AI_PCOD # a folder with all the aipcod.R script and all the Pcod stock synthesis files
# bsaiam.R # Atka mackerel
# goapk.R # GOA pollock

The afscOSA worflow

1. Load afscOSA and data.

2. Structure observed and expected compositions as matrices with nrows = number of years, ncols = number of ages or length bins.

3. Calculate OSA residuals for each fleet using run_osa(). See details for inputs and outputs by running ??run_osa().

4. Plot OSA residuals and aggregate fits for one or more fits using plot_osa(). Input to plot_osa() is a list of output(s) from run_osa(). See more details by running ??plot_osa.

References

Trijoulet V, Nielsen A (2022). compResidual: Residual calculation for compositional observations. R package version 0.0.1.