ratecalib: Calibration Weighting to Multiple Subgroup Pass-Rate Targets

Calibration weighting for binary-outcome pass rates against multiple overlapping subgroup targets. Adjusts initial positive weights so that the overall pass rate and subgroup pass rates approach (soft mode) or exactly match (exact mode) given targets, while preserving the initial weight structure and population margins. Provides a one-step interface, pre-solve data checks, target-table construction, effective sample size and design-effect diagnostics, and example data. The solver works on a bounded convex quadratic program over demographic-cell-by-outcome aggregates for efficiency on large samples. Methods follow the calibration approach of Deville and Saerndal (1992) <doi:10.1080/01621459.1992.10475217>.

Version:	0.3.0
Depends:	R (≥ 4.1.0)
Imports:	Matrix, methods, osqp, stats
Suggests:	testthat (≥ 3.0.0), openxlsx
Published:	2026-06-25
DOI:	10.32614/CRAN.package.ratecalib (may not be active yet)
Author:	Kunxiang Ma [aut, cre]
Maintainer:	Kunxiang Ma <mkx07080412 at gmail.com>
BugReports:	https://github.com/makunxiang-cmd/ratecalib/issues
License:	MIT + file LICENSE
URL:	https://github.com/makunxiang-cmd/ratecalib
NeedsCompilation:	no
Materials:	NEWS
CRAN checks:	ratecalib results

Documentation:

Reference manual:

ratecalib.html , ratecalib.pdf

Downloads:

Package source:	ratecalib_0.3.0.tar.gz
Windows binaries:	r-devel: ratecalib_0.3.0.zip, r-release: not available, r-oldrel: not available
macOS binaries:	r-release (arm64): ratecalib_0.3.0.tgz, r-oldrel (arm64): ratecalib_0.3.0.tgz, r-release (x86_64): ratecalib_0.3.0.tgz, r-oldrel (x86_64): ratecalib_0.3.0.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=ratecalib to link to this page.