The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

gkwdist 1.0.7

gkwdist 1.0.5

Documentation Improvements

• Enhanced Examples for Likelihood Functions: All ll*, gr*, and hs* functions now include comprehensive examples demonstrating:

  • Maximum likelihood estimation with analytical gradients
  • Univariate profile likelihoods with confidence thresholds
  • 2D likelihood surfaces with confidence regions (90%, 95%, 99%)
  • Confidence ellipses with marginal intervals for parameter pairs
  • Numerical vs analytical derivative verification
  • Likelihood ratio tests and score tests

• Professional Visualization Standards:

  • Consistent color scheme across all examples
  • Grid-adaptive algorithms for computational efficiency
  • Base R only - no external dependencies required

• Complete Coverage: Enhanced documentation for all distribution families (Kw, EKw, KKw, GKw) covering 2 to 5 parameters

• Theoretical References: Documentation cites foundational work by Carrasco et al. (2010), Jones (2009), Kumaraswamy (1980), and standard inference theory from Casella & Berger (2002)

gkwdist 1.0.3

• README.md: Fix typos and faill link
  • Fix zzz.R file by removing useless texts

gkwdist 1.0.2

gkwdist 1.0.1

Major Improvements

Enhanced gkwgetstartvalues() Function

• NEW: Added family parameter to support all distribution families
  • Automatically returns correct number of parameters for each family
  • Family-specific initial value strategies for better convergence
  • Supported families: "gkw", "bkw", "kkw", "ekw", "mc", "kw", "beta"
  • Case-insensitive family names for user convenience

Documentation Enhancements

• README.md: Complete rewrite with mathematical rigor
  • All LaTeX formulas corrected and verified for proper rendering
  • Eight comprehensive examples using optim() with analytical gradients
  • Corrected function signatures: all ll*(), gr*(), and hs*() functions use (par, data) signature
  • Added performance benchmarks demonstrating 10-50× speedup with C++ implementation
  • Hierarchical structure diagram for all distribution families
  • Model selection workflow and practical guidelines
  • Removed all references to deprecated gkwfit() function

CRAN Submission Readiness

• DESCRIPTION: Fixed to meet CRAN requirements
  • Proper Authors@R field formatting
  • Removed unused dependencies (numDeriv)
  • Corrected package dependencies (RcppArmadillo only in LinkingTo)
  • Enhanced description with DOI references
  • Fixed maintainer email formatting

Bug Fixes

• Fixed function call signatures in all README examples to match actual implementation
• Corrected parameter passing in optimization examples (now consistently use (par, data))
• Fixed LaTeX rendering issues with \left/\right delimiters in GitHub Markdown

Testing

• NEW: Comprehensive test suite using testthat
  • 100+ tests covering all exported functions
  • Tests for all 7 distribution families (GKw, BKw, KKw, EKw, MC, Kw, Beta)
  • PDF, CDF, quantile, and random generation tests
  • Log-likelihood, gradient, and Hessian validation
  • Parameter recovery tests with MLE
  • Edge cases and boundary condition handling
  • Integration tests for PDF-CDF consistency

Performance

• All functions implemented in C++ for maximum computational efficiency
• Analytical derivatives (gradient and Hessian) provide exact computations
• Optimized numerical stability for extreme parameter values

Notes

• This is the initial CRAN submission
• Package focuses exclusively on distribution functions (no high-level fitting interface)
• Companion package gkwreg provides regression modeling capabilities
• All user-facing functions maintain backward compatibility
• C++ implementation uses RcppArmadillo for linear algebra operations
• Analytical functions use robust log-scale computations to prevent overflow/underflow
• Random generation uses inverse CDF method where closed-form solutions exist

gkwdist 0.1.0

New Features

• Initial CRAN release
• Generalized Kumaraswamy distribution (5 parameters)
• Six nested sub-families: Beta, Kumaraswamy, Exponentiated-Kumaraswamy, Kumaraswamy-Kumaraswamy, Beta-Kumaraswamy, and McDonald distributions
• Complete set of distribution functions (d/p/q/r)
• Log-likelihood, gradient, and Hessian functions for all families

Performance

• Optimized C++ implementation via Rcpp
• Vectorized operations for speed

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.