Skip to contents

Changelog

Source: NEWS.md

gkwreg 2.1.18

Resubmission addressing the CRAN pre-test feedback on 2.1.17: the vignette took about 11 minutes to rebuild, which is more than CRAN can afford to run regularly.

Vignette build time

  • The Monte Carlo study is now precomputed. The study itself is unchanged – still 3 scenarios x 200 replications x 4 models – but the summaries are stored in inst/extdata/vignette-simulations.rds and read by the vignette instead of being recomputed on every build. The generating script is data-raw/vignette-simulations.R.
  • The illustrative model fits stay live, so the vignette still exercises the package; they use a single family, which cuts the run-time TMB compilation from three models to one.
  • Dropped cache = TRUE from the vignette: with the study precomputed it buys nothing, and it was what produced the stray cache directory removed in 2.1.17.

Local vignette build time drops from about 3m30s to 35s.

Fixed

  • Corrected a measurement artefact in the simulation study. Timings were taken around each fitting call, so the one-off TMB compilation (~30s per family) fell inside the first replicate and was then averaged over all 200, inflating every reported gkwreg timing by roughly 0.15s. Scenario 1 consequently reported Kumaraswamy as slower than betareg while the text claimed it was faster. The models are now compiled before the timed loops.
  • Figures quoted in the vignette prose (speed-ups, convergence rates, AIC) are now computed inline from the result tables rather than written out by hand, so the narrative cannot drift away from the results it describes.

gkwreg 2.1.17

Maintenance release. No changes to the statistical methods or to the exported API.

Version 2.1.16 was used for the JOSS review archive on Zenodo and was never released on CRAN; 2.1.17 is the CRAN update that follows 2.1.14.

Publication

  • The methodology and software are now published in the Journal of Open Source Software: Lopes and Bonat (2026), https://doi.org/10.21105/joss.08991.
  • Added inst/CITATION, so that citation("gkwreg") returns the peer-reviewed reference, and added that reference to the Description field.

Dependencies

Fixed

  • Restored genuine UTF-8 characters in the documentation and examples. Accented author names and mathematical symbols had regressed to literal <U+XXXX> escape sequences, affecting the LossAversion, ReadingSkills, gkwreg, anova.gkwreg and residuals.gkwreg help pages, the README and the vignette.
  • Normalised DESCRIPTION, LICENSE, NAMESPACE and all package sources to LF line endings.
  • Restored LICENSE to the two-line DCF stub that License: MIT + file LICENSE requires. It had been replaced by the full MIT text, which R CMD check reports as “License stub is invalid DCF”. The full text remains in LICENSE.md for GitHub.
  • Fixed a typo in utils::globalVariables(), where "dkw dmc" was a single string instead of two separate entries.

Packaging

  • The knitr cache directory of the vignette is no longer under version control and no longer reaches the source tarball; the vignette is always rebuilt from scratch.
  • Tightened .Rbuildignore and .gitignore so that build artefacts, session files and cache directories cannot leak into the tarball.

Documentation

  • README: added the JOSS badge, removed a stale BibTeX block that advertised an outdated version and omitted the second author, and fixed the author footer, which was being rendered as a broken table.
  • Reworked the pkgdown site (light theme, KaTeX math rendering).

gkwreg 2.1.14

CRAN release: 2026-01-09

Fixed

  • clang-san runtime error (integer overflow). Fixed a static_cast<int> overflow in the cache-key generation used by the TMB models. A safe_int_cast() helper now prevents undefined behaviour when distribution parameters reach extreme values during optimisation. Affects gkwreg.cpp, bkwreg.cpp, kkwreg.cpp, ekwreg.cpp, mcreg.cpp and kwreg.cpp.

gkwreg 2.1.13

CRAN Resubmission

Addresses all remaining issues for CRAN acceptance after archival on 2025-11-30.

Fixed

  • Added inst/WORDLIST with ‘Kumaraswamy’ to resolve spelling NOTE
  • Added skip_on_cran() to all test files to reduce check time from 18min to ~9min
  • Added cran-comments.md documenting changes since archival (excluded from build via .Rbuildignore)

Confirmed

  • Cache policy now fully compliant: uses only tempdir() for session-specific TMB DLL cache
  • All ~/.cache/gkwreg usage completely removed (fixed in v2.1.11)
  • Check time now under 10 minutes

gkwreg 2.1.12

CRAN Compliance Fixes

This release addresses all issues that led to package archival on 2025-11-30.

Fixed

  • CRITICAL: Removed RcppArmadillo and RcppEigen from Imports field in DESCRIPTION (they remain in LinkingTo only, as they are used solely for C++ compilation)
  • Added single quotes around technical term ‘Kumaraswamy’ in DESCRIPTION
  • Removed unnecessary @import RcppArmadillo roxygen directive
  • Regenerated NAMESPACE to reflect correct imports

Note

The cache policy violation (~/.cache/gkwreg) reported in version 2.1.6 was already fixed in version 2.1.11. The package now uses ONLY tempdir() for session-specific temporary cache, fully complying with CRAN policies.

gkwreg 2.1.4

Minors and documentation

  • Fixed the link to the LICENSE.md file.
  • Corrected the bold-faced sentences in paper.md.
  • Added contribution guidelines.
  • Fix README.md equations and other small text issues

Other changes:

  • Added new subsections: Distributional Regression Framework and Model Diagnostics in paper.md. This makes the paper more comprehensive.
  • Removed mentions of the distribution family, as those implementations are now in the new package “gkwdisst”.

gkwreg 2.1.1

CRAN release: 2025-11-15

gkwreg 2.1.0

Comparative Testing

  • Introduced a dedicated comparative test suite to validate gkwreg’s beta family implementation against the reference betareg package, ensuring numerical accuracy and reliability.

  • Confirmed statistical equivalence despite different internal parameterizations. Tests demonstrate that gkwreg’s shape-based (gamma, delta+1) approach produces equivalent statistical models to betareg’s mean-precision (mu, phi) approach.

  • Validated key outputs, showing that log-likelihood, AIC/BIC, fitted values, and predictions are virtually identical between the two packages when fitting the same beta regression model.

  • Successfully benchmarked gkwreg with family = "beta" as a robust and reliable alternative for beta regression, yielding the same inferential conclusions as the established betareg package.

  • Verified consistency across multiple scenarios, including controlled simulations with known parameters and real-world datasets (GasolineYield, FoodExpenditure), ensuring robust performance in diverse modeling contexts.

gkwreg 2.0.0

Major Changes

Package Restructuring

  • Complete package reformulation following JOSS reviewer feedback to reduce complexity and improve maintainability.

  • Distribution functions moved to separate package gkwdist: All d*, p*, q*, r* density/CDF/quantile/random generation functions have been extracted to the companion package gkwdist for cleaner namespace and reduced dependencies. The gkwreg package now focuses exclusively on regression modeling.

  • Univariate fitting functions removed: gkwfit(), gkwgof(), and gkwfitall() have been removed to maintain package focus on regression. Users needing univariate distribution fitting should use the gkwdist package directly or standard MLE approaches.

Simplified Interface

  • Introduced gkw_control(): All technical/optimization parameters (method, start, fixed, hessian, maxit, tolerances, etc.) are now consolidated in a dedicated control function following the glm.control() design pattern. This dramatically simplifies the main gkwreg() interface.

  • Removed arguments violating separation of concerns from gkwreg():

    • plot argument removed (use plot() method instead)
    • conf.level argument removed (use confint() method instead)
    • profile, submodels, npoints arguments removed (focused functionality)

  • Streamlined gkwreg() signature: Reduced from 15+ arguments to ~12 core arguments, with technical options delegated to control.

Complete S3 Method Implementation

Enhanced Diagnostics

  • Comprehensive plot.gkwreg() method with 6 diagnostic plot types:

    1. Residuals vs Observation Indices
    2. Cook’s Distance
    3. Generalized Leverage vs Fitted Values
    4. Residuals vs Linear Predictor
    5. Half-Normal Plot with Simulated Envelope
    6. Predicted vs Observed Values

  • Dual graphics system support: Base R graphics (default) or ggplot2 with automatic grid arrangement via gridExtra/ggpubr.

  • Advanced customization: Named-list interface for plot captions (partial customization without repeating all titles), theme control, sampling for large datasets.

Powerful Prediction

  • predict.gkwreg() with 9 prediction types:
    • "response", "variance", "link", "parameter"
    • Individual parameters: "alpha", "beta", "gamma", "delta", "lambda"
    • Distribution functions: "density", "probability", "quantile"
  • Element-wise and vectorized modes: Flexible evaluation via elementwise argument for CDF/PDF/quantile calculations.

Model Comparison Tools

  • Likelihood ratio tests: anova.gkwreg() for comparing nested models with automatic ordering and chi-squared tests; dedicated lrtest() function for pairwise comparisons.

  • Information criteria: AIC.gkwreg() and BIC.gkwreg() with multi-model comparison support returning data frames.

Documentation Improvements

  • Extensive Roxygen documentation for all exported functions with detailed examples, mathematical formulas, and usage guidance.

  • Updated README.md with comprehensive feature overview, quick start guide, advanced examples, and ecosystem comparison table.

  • NULL default intelligent behavior: Several arguments default to NULL triggering smart auto-configuration (e.g., sub.caption, ask, theme_fn in plot.gkwreg()).

Testing Framework

Comprehensive Test Suite Added

The package now includes a robust testing framework with 1000+ unit tests covering all major functionalities:

Core Function Testing
  • gkwreg(): 20 tests for model fitting, parameter estimation, formula handling, all distribution families, link functions, and convergence
  • predict.gkwreg(): 10 tests for predictions, including response means, densities, CDFs, quantiles, and parameter extraction
  • residuals.gkwreg(): 10 tests for all residual types (response, Pearson, deviance, quantile, standardized, working, partial)
  • fitted.gkwreg(): 10 tests for fitted value extraction and validation
S3 Methods Testing
Test Coverage Includes
  • ll 7 distribution families (GKw, BKw, KKw, EKw, MC, Kw, Beta)
  • Different link functions and scales
  • Edge cases and boundary conditions
  • Missing data handling (NA)
  • Subset and weight specifications
  • Large dataset performance
  • Error handling and input validation
  • Statistical correctness verification
  • Numerical accuracy checks
Testing Framework
  • Built with testthat package
  • Uses simulated data from gkwdist package
  • Tests with real datasets (GasolineYield, FoodExpenditure)
  • Reproducible with fixed random seeds

Minor Improvements

  • Link scaling support: Added link_scale argument to gkwreg() for controlling transformation intensity.

  • Performance optimizations: Intelligent caching, sampling support for diagnostics on large datasets, optional Hessian computation.

Breaking Changes: Version 2.0.0 introduces breaking changes. Code using gkwfit(), gkwgof(), gkwfitall(), or distribution functions (dgkw(), etc.) must be updated to use the gkwdist package or the new gkwreg() interface with gkw_control().