This is a fork of https://github.com/fishfollower/SAM, where we model the age-dependent
parameters logFpar and logSdLogObs using smoothing splines instead of the standard manual
grouping method. The model extension is described in the paper "Adding smoothing splines to the SAM
model improves stock assessment".
The package can be installed by calling make from your terminal, when located inside this directory.
All code used to create the results in the paper is available in the folder scripts/. This
folder consists of three scripts. The first script, model_fitting.R, is used for fitting four
competing versions of the SAM model to various different fish stock data sets. Cross-validation and
forward-validation studies are also performed in this script. The second script,
examination_of_results.R, is used to examine the results of all the model fitting performed in
model_fitting.R. Finally, the script helper_functions.R consists of a set of helper functions
that are used in the two previously mentioned scripts.
Recent changes were made to the SAM package while we were developing our spline models. These recent
changes have been merged into this fork, and minor changes have been made to the code so that the
spline models will be usable with the newest version of SAM. However, the actual version of SAM that
was used for creating all the results in the paper, can be found by checking out the tag
"SAM-spline-paper", i.e. git checkout tags/SAM-spline-paper.