Skip to contents

Changelog

Source: NEWS.md

AgeingError 2.0.2

  • Switched to the new TMB version of the package in the main branch
  • Renamed github repository from nwfscAgeingError to AgeingError
  • Adding vignette documenting the TMB version from Paul Burch
  • Cleared out some, but not all, of the files associated with the ADMB version

AgeingError 2.0.0

  • Changed package name to AgeingError
  • Added TMB code that is pre-compiled in the package
  • Added run() as a wrapper for the code provided by @puntae
  • Increased the number of examples present in the package

nwfscAgeingError 1.3.3

  • Added a NEWS.md file to track changes to the package.
  • Included TMB agemat2.cpp source code from A. E. Punt

nwfscAgeingError 1.3.2

This version contains updates by A. E. Punt during his work with CSIRO.

nwfscAgeingError 1.3.1

Archive from 16th of November 2021.

nwfscAgeingError 1.0.1

The version used for the 2017 assessment cycle.

nwfscAgeingError 1.0.0

The version used for the 2013 assessment cycle. Though the repository was not tagged until 2015. Melissa Haltuch noted that in 2014 the code was updated and a summary of those updates are provided below.

  • tpl file
    • 2 new options for bias estimation.
        1. Splines, which requires a new argument defining the knots.
        1. Linear interpolation between knots, which requires a new argument that defines the knots.
    • Changes the calculation of the best true age because the previous way led to strange plots.
    • Improvement to effective sample size calculation.
  • dat file
    • Read in one data set with missing entries, where there are no double reads. In the 2011-based code it allowed for reading in of multiple data sets.
    • Increased estimation speed because of how missing values are treated.
  • R files
    • Includes code to simulate data.
    • Includes a function to format input data from a csv with readers in columns and double reads in the rows.
    • Plot standardized outputs from the output file. These figures could still use some development.
    • Ability to step through multiple runs and catalog the output using the new StepwiseFn() function.