ilovemaths/spfcICOMP: spfcICOMP0.1.0

Abstract

spfcICOMP 0.1.0 This is the first CRAN release of spfcICOMP, an R package for shrinkage principal fitted components with information complexity-based model selection in high-dimensional regression and classification. Main capabilities Regularised covariance estimation using Oracle Approximating Shrinkage and Maximum Entropy Covariance Response-guided feature screening for high-dimensional predictors Structural-dimension selection using conventional and information-complexity criteria Reduced-space regression and classification Train-test splitting and cross-validation utilities Simulation tools for assessing dimension recovery, feature screening and predictive performance Installation Install the released version directly from CRAN: install.packages("spfcICOMP") library(spfcICOMP) Release verification Version 0.1.0 completed the following checks: 149 of 149 test expectations passed GitHub Actions checks passed on macOS, Windows, Ubuntu R-release, Ubuntu R-devel and Ubuntu R-oldrel Compatibility with the declared requirement of R 3.5.0 or later was audited CRAN incoming pretests completed with no errors or warnings The package was published on CRAN on 3 September 2026 Links CRAN package: https://cran.r-project.org/package=spfcICOMP Documentation: https://cran.r-project.org/web/packages/spfcICOMP/ Reference manual: https://cran.r-project.org/web/packages/spfcICOMP/spfcICOMP.pdf Source repository: https://github.com/ilovemaths/spfcICOMP Issue tracker: https://github.com/ilovemaths/spfcICOMP/issues Citation Please use the package citation supplied by R: citation("spfcICOMP") The release translates methodological work on shrinkage covariance estimation, sufficient dimension reduction and information-complexity-based model selection into documented and reproducible statistical software.