Package: logistf Version: 1.26.1 Date: 2025-04-16 Title: Firth's Bias-Reduced Logistic Regression Authors@R: c(person("Georg", "Heinze", role=c("aut", "cre"), email="georg.heinze@meduniwien.ac.at"), person("Meinhard", "Ploner", role=c("aut")), person("Daniela","Dunkler", role=c("ctb")), person("Harry", "Southworth", role=c("ctb")), person("Lena", "Jiricka", role=c("aut")), person("Gregor", "Steiner", role=c("aut"))) Depends: R (>= 3.0.0) Imports: mice, mgcv, formula.tools, Matrix Suggests: emmeans (>= 1.4), estimability Description: Fit a logistic regression model using Firth's bias reduction method, equivalent to penalization of the log-likelihood by the Jeffreys prior. Confidence intervals for regression coefficients can be computed by penalized profile likelihood. Firth's method was proposed as ideal solution to the problem of separation in logistic regression, see Heinze and Schemper (2002) . If needed, the bias reduction can be turned off such that ordinary maximum likelihood logistic regression is obtained. Two new modifications of Firth's method, FLIC and FLAC, lead to unbiased predictions and are now available in the package as well, see Puhr et al (2017) . License: GPL URL: https://github.com/georgheinze/logistf LazyLoad: yes NeedsCompilation: yes RoxygenNote: 7.2.2 LazyData: true BugReports: https://github.com/georgheinze/logistf/issues/ Roxygen: list(markdown = TRUE) Config/pak/sysreqs: cmake make libicu-dev libx11-dev zlib1g-dev Repository: https://georgheinze.r-universe.dev Date/Publication: 2025-04-16 15:01:54 UTC RemoteUrl: https://github.com/georgheinze/logistf RemoteRef: HEAD RemoteSha: f9f36414c3124a3bc65b0b690d29ca581e080438 Packaged: 2026-05-31 07:27:21 UTC; root Author: Georg Heinze [aut, cre], Meinhard Ploner [aut], Daniela Dunkler [ctb], Harry Southworth [ctb], Lena Jiricka [aut], Gregor Steiner [aut] Maintainer: Georg Heinze