Skip to content

Releases: ASKurz/Statistical_Rethinking_with_brms_ggplot2_and_the_tidyverse_2_ed

expand_grid() and linewidth

26 Jan 17:54
Compare
Choose a tag to compare

Noteworthy changes include:

  • replacing my incorrect use of tidyr::expand() with a more appropriate tidyr::expand_grid() workflow, and
  • adopting the new linewidth argument for several ggplot2 geoms.

as_draws_df()

29 Sep 16:08
32f14ff
Compare
Choose a tag to compare

Major improvements include:

  • an as_draws_df()-oriented workflow to replace the depreciated posterior_samples() function;
  • better sampling in models in the first few due to changes in the ub argument;
  • more use of the uniform prior for σ to better match the text, thanks to the improved ub argument;
  • a more tidyverse-oriented version of the sim_happiness() function in Section 6.3, thanks to Randall Pruim
  • a more accurate depiction of entropy in Figure 10.2.b in Section 10.1.1 thanks to help from Hamed Bastan-Hagh,
  • a cross link to the first edition, thanks to a suggestion from Will Petry;
  • some corrected typos and updates to the links and citations; and
  • all models have been refit using brms version 2.18.0.

99% there

16 Mar 19:35
19a4c5f
Compare
Choose a tag to compare

Major improvements include:

  • a corrected workflow for fitting single-level b-spline models (Section 4.5.2), thanks to Stephen Wild;
  • a refined workflow for fitting multilevel b-spline models using the s() function (Section 4.6), thanks to Gavin Simpson;
  • a new bonus section (4.7) on grouping predictors within matrix columns, thanks to insights from Gelman et al (2020) and hints from Paul-Christian Bürkner;
  • a brms solution to the sixth-order polynomial model in Section 7.1.1, thanks again to Stephen Wild;
  • a workflow to reproduce the Metropolis simulation for Figure 9.3, thanks to James Henegan;
  • a corrected workflow for taking fitted()-based random draws for the bonus material in Section 13.5.2.1, thanks to an exchange with Ladislas Nalborczyk;
  • a brms solution to McElreath's bivariate differential equation model in Section 16.4, thanks to Markus Gesmann;
  • better use of geom_area() throughout the book, thanks to insights from Randall Pruim;
  • the adoption of a CONTRIBUTING section on GitHub, thanks to Brenton M. Wiernik;
  • improved table workflow with the flextable package; and
  • all models have been refit using brms version 2.15.0.

broom, slice_sample(), and typos

02 Dec 20:08
Compare
Choose a tag to compare

This is a mini update designed to

  • fix code breaks resulting from updates to the broom package, caught by Jenny Bigman;
  • replace the soon-to-be retired sample_n() code with slice_sample(), caught by Randall Pruim;
  • and correct a few typos along the way.