Package: rema 1.2.0

Jane Sullivan

rema: A generalized framework to fit the random effects (RE) model, a state-space random walk model developed at the Alaska Fisheries Science Center (AFSC) for apportionment and biomass estimation of groundfish and crab stocks.

This package provides a generalized framework to fit the random effects (RE) model, a state-space random walk model developed at the Alaska Fisheries Science Center (AFSC) for smoothing survey biomass estimates and apportioning catch among management areas. REMA is a multivariate extension of the original single-survey, single-strata RE model that allows the use of multiple strata within a survey and an additional survey (e.g. CPUE or relative population numbers) to inform the biomass trend (Hulson et al. 2021). If multi-survey mode is turned off, REMA runs the same as the univariate (RE) and multivariate (i.e. multiple area or depth strata; REM) versions of the model. REMA was developed in Template Model Builder (TMB; Kristensen et al. 2016).

Authors:Jane Sullivan [aut, cre], Laurinne Balstad [aut, ctb], Cole Monnahan [ctb], Pete Hulson [ctb]

rema_1.2.0.tar.gz
rema_1.2.0.zip(r-4.7-x86_64)rema_1.2.0.zip(r-4.6-x86_64)rema_1.2.0.zip(r-4.5-x86_64)
rema_1.2.0.tgz(r-4.6-x86_64)rema_1.2.0.tgz(r-4.6-arm64)rema_1.2.0.tgz(r-4.5-x86_64)rema_1.2.0.tgz(r-4.5-arm64)
rema_1.2.0.tar.gz(r-4.7-arm64)rema_1.2.0.tar.gz(r-4.7-x86_64)rema_1.2.0.tar.gz(r-4.6-arm64)rema_1.2.0.tar.gz(r-4.6-x86_64)
rema_1.2.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION
card.svg |card.png
rema/json (API)

# Install 'rema' in R:
install.packages('rema', repos = c('https://jimianelli.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/janesullivan-noaa/rema/issues

Pkgdown/docs site:https://afsc-assessments.github.io

Uses libs:
  • c++– GNU Standard C++ Library v3

On CRAN:

Conda:

cpp

4.95 score 9 stars 40 scripts 289 downloads 13 exports 58 dependencies

Last updated from:9d80130009. Checks:11 WARNING, 2 OK. Indexed: no.

TargetResultTimeFilesSyslog
linux-devel-arm64WARNING213
linux-devel-x86_64WARNING214
source / vignettesOK345
linux-release-arm64WARNING251
linux-release-x86_64WARNING224
macos-release-arm64WARNING175
macos-release-x86_64WARNING446
macos-oldrel-arm64WARNING156
macos-oldrel-x86_64WARNING344
windows-develWARNING192
windows-releaseWARNING185
windows-oldrelWARNING197
wasm-releaseOK167

Exports:%>%check_convergencecheck_estimabilitycompare_rema_modelsfit_remaget_osa_residualsplot_extra_cvplot_remaprepare_rema_inputread_admb_reread_reptidy_extra_cvtidy_rema

Dependencies:abindbase64encbslibcachemclicpp11digestdplyrevaluatefarverfastmapfontawesomefsgenericsggplot2gluegtablehighrhtmltoolsisobandjquerylibjsonliteknitrlabelinglatticelifecyclemagrittrMatrixmemoisemimepillarpkgconfigpurrrR6rappdirsRColorBrewerRcppRcppEigenrlangrmarkdownS7sassscalessessioninfostringistringrtibbletidyrtidyselecttinytexTMButf8vctrsviridisLitewithrxfunyamlzoo

REMA model equations
Base model structure for a single survey and stratum | Extending to multiple biomass survey strata | Addition of an auxiliary catch per unit effort (CPUE) survey | The ADMB version of REMA | Estimation of additional observation error | Exploration of the Tweedie distribution for zero biomass observations | Experimental One-Step Ahead (OSA) residuals | References
Rendered fromrema_equations.Rmdusingknitr::rmarkdown

Last update: 2025-01-07
Started: 2022-08-15

REMA model validation
Background | A testing framework for REMA model validation | Prepare data for each stock here | 1. Simulation self-test: Can we recover parameters without bias? | 2. Residual analysis | 3. Laplace approximation: Are the model assumptions related to random effects estimation reasonable? | 4. Parameter correlation: Are the model parameters identifiable and non-redundant? | Results of parameter correlation analysis | 5. Is the GOA Thornyhead model really converged? | When should we care about failed model diagnostics? | Preliminary recommendations | References
Rendered fromex4_model_validation.Rmdusingknitr::rmarkdown

Last update: 2024-12-04
Started: 2024-08-21

Strategies for handling zero biomass observations
Model 1: Zeros as NAs | Model 2: Add a small constant | Model 3: Tweedie distribution | Compare model results
Rendered fromex3_zeros.Rmdusingknitr::rmarkdown

Last update: 2024-08-21
Started: 2022-08-15

Fitting to an additional CPUE survey
Model 1: Fit to the biomass and CPUE survey through the estimation of a scaling parameter $q$ | A comparison of the ADMB and TMB models | Model 2: Estimating additional observation error in the two surveys | Model comparison | Model selection | Apportionment results
Rendered fromex2_cpue.Rmdusingknitr::rmarkdown

Last update: 2023-08-31
Started: 2022-08-15

REMA basics
The rema workflow: | Example 1: Univariate random effects (RE) model with a single survey and stratum | Example 2: Multivariate random effects model (REM) with a single survey and multiple strata
Rendered fromex1_basics.Rmdusingknitr::rmarkdown

Last update: 2023-08-30
Started: 2022-06-16