brglm2 1.0.0

Improvements, updates and additions

• Provides mdyplFit() method for the glm() function, which estimates logistic regression models using maximum Diaconis-Ylvisaker prior penalized likelihood, and associated methods. Corrections to estimates, estimated standard errors, z statistics, and so on, can be applied using the developments Sterzinger and Kosmidis (2024, doi:10.48550/arXiv.2311.07419) by setting hd_correction = TRUE in the calls to the summary() and confint() methods for "mdyplFit" objects.

• Added add1() and drop1() methods for mdyplFit objects.

• Added MultipleFeatures data set, which is a subset of Duin (1998, doi:10.24432/C5HC70), as analyzed in Sterzinger and Kosmidis (2024, doi:10.48550/arXiv.2311.07419).

• Documentation updates and documentation of new methods.

Bug fixes

• Added add1() and drop1() methods for brglmFit objects, so that
  step() does not default to the methods for glm objects (which
  would return nonsense results); thanks to jamiahuswalton@github
  (#33) for reporting the
  issue.

• Sane staring values, graceful failing, updates to optimization
  procedure, and improvements to methods for brnb(); thanks to
  cperk@github (#31) for
  reporting issues with brnb() when infinite ML estimates are
  encountered.

Improvements, updates and additions

• Documentation updates

Improvements, updates and additions

• Convergence of the brglm_fit iterations is now determined if the L^Inf norm of the step size (rather than the L^1 as it was previously) of the quasi-Fisher scoring procedure is less than epsilon (see ?brglm_control for the definition of epsilon). This is more natural as epsilon then determines directly the precision of the reported estimates and does not depend on their number.

• brglm_control() now checks that the supplied value of max_step_factor is numeric and greater or equal to 1. If not, then it is set to the default value of 12.

• Vignette updates

brglm2 0.9.1

Other improvements, updates and additions

• Added the enzymes and hepatitis data sets (from the pmlr) to support examples and tests.

brglm2 0.9.0

New functionality

• The expo() method for brglmFit and glm objects estimates the exponential of parameters of generalized linear models with maximum likelihood or various mean and median bias reduction methods (see ?expo for details). The expo() method is particularly useful for computing (corrected) estimates of the multiplicative impact of a unit increase on a covariate on the mean of a Poisson log-linear model (family = poisson("log") in glm()) while adjusting for other covariates, the odds ratio associated with a unit increase on a covariate in a logistic regression model (family = binomial("logit") in glm()) while adjusting for other covariates, the relative risk associated with a unit increase on a covariate in a relative risk regression model (family = binomial("log") in glm()) while adjusting for other covariates, among others.

Bug fixes

• Fixed a bug where the dispersion in the resulting object would not be transformed even if transformation != "identity" when type is ML or AS_median or AS_mixed.

Other improvements, updates and additions

• Moved unit tests to tinytest.

• Moved documentation to markdown markup through roxygen2.

• New vignette titled "Estimating the exponential of regression parameters using brglm2", to demonstrate the expo() method.

• Various documentation fixes.

brglm2 0.8.2

Other improvements, updates and additions

• Housekeeping.
• Removed lpSolveAPI from imports.

brglm2 0.8.1

Bug fixes

• Fixed a bug when predicting from bracl objects with non-identifiable parameters.

Other improvements, updates and additions

• Work on output consistently from print() methods for summary.XYZ
  objects; estimator type is now printed and other fixes.

• Enriched warning when algorithm does not converge with more informative text.

• Documentation fixes and updates

New functionality

• brnb() allows fitting negative binomial regression models using
  implicit and explicit bias reduction methods. See vignettes for a
  case study.
• simulate() method for objects of class brmultinom and bracl
• ordinal_superiority() method to estimate Agresti and Kateri
  (2017)'s ordinal superiority measures, and compute bias corrections
  for those.

Bug fixes

• Fixed a bug that would return an error when Wald.ratios = TRUE in
  summary.brmultinom.
• Fixed bug in vcov.bracl that would return an error if the
  "bracl" object was computed using bracl() with parallel = TRUE
  and one covariate.
• Fixed a bug in bracl() related to the handling or zero weights
  that could result in hard-to-traceback errors.
• Fixed a bug in bracl() that could cause errors in fits with one
  covariate.
• brglmFit() iteration returns last estimates that worked if
  iteration fails.

Other improvements, updates and additions

• Documentation and example updates.

Bug fixes

• Fixed bug where confint() was not returning anything when applied
  to objects of class brmultinom.
• Fixed bug where and error could result when the control glm().
  argument was specified using the output from brglmControl() or
  brglm_control().

New functionality

• added check_aliasing option in brglmControl() to tell
  brglm_fit() to skip (check_aliasing = TRUE) or not
  (check_aliasing = FALSE) rank deficiency checks (through a QR
  decomposition of the model matrix), saving some computational effort.

Other improvements, updates and additions

• Updated DOI links in documentation and some http -> https fixes.

Bug fixes

• Fixed bug that resulted in NA coefficients when brglmFit was
  called with a vector x or an x with no column names

New functionality

• confint method for brmulitnom objects

Other improvements, updates and additions

• Updated reference to Kenne Pagui et al (2017)
• Updated reference to Kosmidis and Firth (2020)
• Fixed issues with references
• Updated documentation

Other improvements, updates and additions

• vcov.brglmFit objects now uses vcov.summary.glm and supports the complete argument for controlling whether the variance covariance matrix should include rows and columns for aliased parameters.
• Deprecated detect_sepration and check_infinite_estimates, which will be removed from brglm2 at version 0.8. New versions of detect_sepration and check_infinite_estimates are now maintained in the detectseparation R package.
• Fixed typos in NEWS.

Bug fixes

• Fixed bug in AIC reported by print.summary for brmultinom and bracl
• detect_separation now handles one-column model matrices correctly

Other improvements, updates and additions

• Documentation improvements and typo fixes