Added Pmat() and Qmat() to the auxiliary
functions for enriched glm objects. These return the
P_t and Q_t matrices used in mean and median
bias-reduction calculations, and can be evaluated at user-supplied
regression and dispersion parameter values.
Bug fixes
Fixed bugs with the handling aliasing.
Other improvements,
updates, and additions
Added new tests for existing and new functionality.
Various documentation updates.
Updated package metadata, including adding brglm2 to
Suggests.
enrichwith 0.4.0
Other improvements,
updates, and additions
Deprecated enriched_glm() as it might lead to
scoping issues. Users are recommended to use the call
enrich(glm(...)) or
glm(...) |> enrich().
Fixed documentation links, and other minor documentation
improvements
enrichwith 0.3.2
Bug fixes
fixed a typo (dmu.deta -> d2mu.deta)
for betareg enrichment, which would cause failures with
get_bias_function() and other methods.
Other improvements,
updates, and additions
Documentation fixes.
Added MASS in Suggests to allow for using
ggplot2 in vignettes.
enrichwith 0.3.1
Bug fixes
Added conditional use of SuppDists.
Other improvements,
updates and additions
Fixed typos and spelling mistakes in documentation.
Added new tests for d/r/p/qmodel.
enrichwith 0.3
New functionality
Added support for wedderburn family from the
gnm R package.
Other improvements,
updates and additions
Improved documentation.
Improved tests and added some new.
enrichwith 0.2
Bug fixes
Fixed bug in the computation of beta regression score contributions
from betareg objects.
Fixed bug with the attributes of information matrices for
generalized linear models with fixed dispersion.
Fixed the row names in the information matrices for glm
and betareg objects.
New functionality
get_information_function for betareg
objects returns a function that can compute observed information
matrices (by setting type = "observed").
get_information_function now returns a function that
can compute the Cholesky decomposition of the information matrix (by
setting CHOL = TRUE).
simulate methods (coming form
get_simulate_function) now check for correct length of
coefficient vector if this is specified.
enriched_glm can now be used to fit GLMs and get
objects that are enriched with auxiliary functions and other components
when compared to their glm counterparts.
Other improvements,
updates and additions
Fixed typos in ?enrich.family.
Added documentation for get_dmodel_function.glm,
get_pmodel_function.glm,
get_qmodel_function.glm.
enrichwith 0.03
Bug fixes
Took care of aliasing in bias calculations.
Names of score and bias components do not get a prefix anymore.
Fixed a bug on argument of information function (part of the
“auxiliary functions” enrichment option).
Fixed a bug when testing the simulation methods.
New functionality
Provide enrich capabilities for lm objects.
Score functions (?get_score_function) now accept
contributions = TRUE for getting score contributions.
Provide the get_*_function convenience methods
(e.g. get_score_function, get_information_function,
get_bias_function).
Provide the dmodel, pmodel, qmodel auxiliary functions for glm
objects, to compute d, p, q based on a supplied data frame.
Other improvements,
updates and additions
Various codebase improvements.
Various documentation improvements.
New vignette for enriching glm objects.
Updated README file.
enrichwith 0.02
Fixed an issue with the example in ?enrich.glm.
Minor codebase improvements.
Harmonised the output of the functions in the auxiliary_functions
component of enriched glm objects.
More detailed descriptions for the enrichment options for
glm objects.
Included a vignette on how bias-reduction for GLMs can be
implemented using enriched glm objects.