NEWS

---

simmr 0.5.1

• Fixed bug introduced in 0.5.0 affecting the prior distribution on the residual standard deviation
  • Added extra argument to simmr_mcmc and simmr_mcmc_tdf to enable more control on the prior distribution on the residual standard deviation

simmr 0.5.0 (2023-03-25)

• Implemented new Fixed Form Variational Bayes method for fitting SIMMs (publication forthcoming)
  • Added new GGally matrix plots

simmr 0.4.6

• Implemented vdiffr for better checking of output plots
  • Added capture.output to remove verbose testing
  • Added feature to allow for >2 sources to combined in combine_sources
  • Added ability to use data in matrix, data frame, or tibble format without error
  • Changed the way prior_viz object is plotted and returned to allow for greater customisation
  • Improved test that to test for bad source mean and sd shapes
  • Added a new vignette on advanced plotting (and moved other parts out of main vignette)

simmr 0.4.5 (2021-02-27)

• Updated new checkmate error checking for multiple functions
  • Added new tests for 90%+ code coverage
  • Fixed a bug that stopped some plots being outputted correctly
  • Fixed some broken examples

simmr 0.4.4 (2021-01-20)

• Updated the simmr_elicit function to provide a more explicit warning for bad input objects
  • Updated compare_sources and compare_groups to allow exporting of the plot object for editing purposes
  • Used styler to correct code style

simmr 0.4.3

• Fixed a bug in combine_sources that stopped it working for multiple groups
  • Added in a load more tests to increase code coverage
  • Implemented checkmate for error checking in simmr_load

simmr 0.4.2 (2020-06-11)

• Fixed a bug with the summary function which was always reporting the same group when an individual group was specified (didn't apply to multiple group summaries). Added a test for that bug.
  • Updated posterior_predictive to produce some more helpful output

simmr 0.4.1 (2019-07-03)

• Fixed some major bugs to plot.simmr_output and compare_groups which caused the wrong groups to be selected
  • Fixed a minor dependency bug as no longer using coda

simmr 0.4.0 (2019-06-14)

• Included prior_viz to visualise and contrast the prior and posterior distributions
  • Included posterior_posterior_predictive to visualise model fit using bayesplot
  • Added simmr_mcmc_tdf to estimate trophic discrimination factors for known diet studies
  • Updated simmr_mcmc to use R2jags