LoginSign Up
← Back to all sparks
R

robma

ANALYTICS
Velocity0.0

R package robma by fbartos — release notes from GitHub.

github.com/fbartos/robma ↗

RoBMA 4.0 tears out its own constructor surface and rebuilds on one class hierarchy

r-packagemeta-analysisbayesianapi-redesignpublication-bias
◆Current state
RoBMA fits robust Bayesian model-averaged meta-analyses that adjust for publication bias. The 3.x line grew by accretion: separate constructors for each model family (RoBMA.reg, NoBMA, BiBMA and their .reg variants), a spike-and-slab algorithm in 3.3.0 that made estimation fast enough to matter, then a steady stream of post-estimation tooling gated on that algorithm — heterogeneity summaries, residuals, funnel plots, z-curve conversion, predict, extract, pooled and adjusted effects. Version 4.0.0 in May 2026 collapses all of it into a unified brma class hierarchy.
◆Where it's heading
The 3.x series solved the modeling problem and left an interface problem behind: a caller had to know which of six constructors matched their data type, and argument names differed across them. 4.0.0 resolves that by making the model family a set of arguments rather than a function name, and by standardizing input naming on metafor-style conventions. It shipped one day after BayesTools 0.3.0, the author's own upstream infrastructure package, whose new standardization and prior-transformation machinery this rewrite depends on.
◆Prediction
A rewrite this wide usually needs a follow-up, so expect 4.0.x patches addressing migration gaps as users hit the removed constructors and renamed arguments.

◆Recent moves

  1. 3mo ago

    Unifies six model constructors into one brma class hierarchy

    ⚡ SPARK

    Rewrites the package around a unified brma class hierarchy, removing the legacy RoBMA.reg, NoBMA, NoBMA.reg, BiBMA and BiBMA.reg constructors and replacing the input alias vocabulary with metafor-style names. The interface debt accumulated across the whole 3.x series gets paid off in one breaking release.

    View source ↗
  2. 8mo ago

    RoBMA 3.6.1

    Adds navigational and worked-example vignettes, a publication bias model type summary, and finer control over z-curve plot elements. Documentation and presentation polish on the 3.6 feature set, immediately before the 4.0 rewrite.

    View source ↗
  3. 11mo ago

    RoBMA 3.6.0

    Adds funnel plots, a residuals() method, and as_zcurve() to reshape meta-analytic models into z-curve objects, all gated on the spike-and-slab algorithm. Another instalment in the pattern of diagnostic tooling arriving only for the faster estimation path.

    View source ↗
  4. 1y ago

    RoBMA 3.5.1

    Adds predict() and true_effects() methods, an extract() accessor for posterior samples, and a standardized_coefficients toggle on summary(). Opening up the posterior for downstream use is the same instinct that later drives the 4.0 interface unification.

    View source ↗
  5. 1y ago

    RoBMA 3.5.0

    Adds computationally feasible three-level selection models via study_ids, three-level binomial-normal models for binary data, and pooled_effect() and adjusted_effect() helpers. It also rescales the PEESE prior, a change that alters estimates rather than just adding options.

    View source ↗
  6. 1y ago

    RoBMA 3.4.0

    Extends binomial-normal meta-regression to binary data via BiBMA.reg and brings the spike-and-slab algorithm to BiBMA models. Each family gaining its own constructor here is exactly the growth pattern 4.0.0 later reverses.

    View source ↗