NEWS

---

subsampling 0.4.0 (2026-06-21)

• Expanded ssp.glm.rF() with a more comprehensive rare-feature subsampling implementation, including additional criteria, automatic rare-feature detection, response balancing options, and union-based combination.
• Improved summaries for rare-feature GLMs and rare-event logistic regression, including clearer sample-composition information.
• Added broader tests for rare-feature GLM scenarios and updated rare-event logistic regression summary tests.
• Revised package vignettes, including the rare-feature GLM vignette and wording cleanup across method vignettes.

subsampling 0.3.0 (2026-03-10)

• Added ssp.glm.rF(). It provides balanced subsampling methods to deal with rare features in GLMs.

subsampling 0.2.0

Changes to ssp.glm()

• Unified the representation of weights in the square score and information matrices to improve readability.
• Added support for family = "gaussian".
• Updated the default value of alpha from 0 to 0.01.
• Added a new output element comp.time that reports computation time.

subsampling 0.1.1 (2024-11-05)

subsampling 0.1.0

• Initial CRAN submission.

• First Version submitted to CRAN.

• After this submission, all updates will be logged here.