lgspline - v0.2.1 - CRAN Resubmission

CRAN Resubmission

After release v0.2.0 was accepted by CRAN (and subsequently released as v0.1.0 on CRAN), it was found to fail on macos-arm64 and Linux Fedora with clang compiler.

This only affected test-advanced.R which failed.

test-advanced.R was updated to improve computational stability and catch situations where model returns a try-error instead of expected output.

rhub check was performed for macos-arm64.

No functionality was adjusted.

- R
Published by matthewlouisdavisBioStat about 1 year ago

lgspline - v0.2.0 - CRAN Submission Ready

lgspline v0.1.0 - CRAN Resubmission

This is a resubmission addressing all points raised by CRAN reviewers and other minor issues.

Changes in this version:

1. Acronym explanation

• Expanded "AFT" to "accelerated failure time (AFT)" throughout the package
• Updated all Weibull AFT-related helper function documentation

2. References in DESCRIPTION

• Added properly formatted references with DOIs and ISBNs for:
  • Ezhov et al. (2018)
  • Searle et al. (2009)
  • Nocedal & Wright (2006)
  • Wahba (1990)
  • Wood (2006)
• Added WORDLIST to /inst/ folder for spell checks

3. Added missing \value tags

• Added detailed return value descriptions to all relevant documentation

4. Fixed example code issues

• Replaced "dontrun" with "donttest" for longer examples
• Unwrapped examples executable in < 5 seconds

5. Fixed par() reset in examples

6. Additional Documentation Improvements

• Standardized notation across examples
• Fixed function calls and parameter descriptions
• Multiple clarity improvements

7. Function Improvements

• Fixed scaling error in polynomial expansions
• Improved predict.lgspline to use fitting data by default when no new data is provided

R CMD check results

0 errors | 0 warnings | 0 notes

- R
Published by matthewlouisdavisBioStat about 1 year ago

lgspline - lgspline 0.1.0: Initial CRAN Release

Initial release of the lgspline package for submission to CRAN.

Key features: - Implements and interpretable variant of smoothing splines - Supports generalized linear models, Weibull accelerated failure time (AFT) models, inequality constraints with quadratic programming, and arbitrary correlation structures - Provides tools for fitting, prediction, and inference - Provides functionality for performing Bayesian Optimization - Options for parallel processing

This package offers a comprehensive framework for fitting smooth function estimations with explicit interpretability of basis functions.

- R
Published by matthewlouisdavisBioStat about 1 year ago