broom 0.7.10

• Clarifies error when pysch::mediate output is dispatched to tidy.mediate (#1037 by @LukasWallrich).
• Allows user to specify confidence level for tidy.rma (#1041 by @TarenSanders)
• Clarifies documentation related to usage of augment_columns(); most package users should use augment() in favor of augment_columns(). See ?augment_columns for more details.
• Extends support for emmeans by fixing non-standard column names in case of asymptotically derived inferential statistics. (#1046 by @crsh)
• Fixes use of index columns in augment.mlogit and adds .resid column to
  output. (#1045, #1053, #1055, and #1056 by @jamesrrae and
  @gregmacfarlane)
• Correct column naming of standard error measures in glance.survfit().
• Various bug fixes and improvements to documentation.

broom 0.7.9

• Fix confidence intervals in tidy.crr(), which were previously exponentiated when exponentiate = FALSE (#1023 by @leejasme)
• Update ergm tidiers in anticipation of changes in later releases. (#1034 by @krivit)
• Address breakages in lfe and Rchoice tidiers

broom 0.7.8

• Fixed bug in glance.ergm related to handling of MCMC details.
• Address breakages in unit tests for {fixest} tidiers.

broom 0.7.7

• Fixed bug in tidy.epi.2by2 that resulted in errors with new version of epiR (#1028 by @nt-williams)
• Added exponentiate argument to tidy.gam() tidier applicable for parametric terms (#1013 by @ddsjoberg)
• Added exponentiate argument to tidy.negbin() tidier (#1011 by @ddsjoberg)
• Fixed failures in spdep tidiers following breaking changes in the most recent release
• Various bug fixes and improvements to documentation

broom 0.7.6

• Fixed bug in augment tidiers resulting in .fitted and .se.fit array columns.
• Fixed bug that made column y non-numeric after tidy_xyz (#973 by @jiho)
• Added tidiers for MASS:glm.nb (#998 by @joshyam-k)
• Fixed bug in tidy.fixest that sometimes prevented arguments like se from being used (#1001 by @karldw)
• Fixed bug in tidy.fixest that resulted in errors when columns with name x are present (#1007 by @grantmcdermott)
• Moved forward with planned deprecation of gamlss tidiers in favor of those provided in broom.mixed
• Various bug fixes and improvements to documentation

broom 0.7.5

• Fixed bug in the nnet::multinom tidier in the case that the response
  variable has only two levels (#993 by @vincentarelbundock and @hughjonesd)
• Small fixes in anticipation of the upcoming survival release
• Various bug fixes and improvements to documentation

broom 0.7.4

broom 0.7.4 introduces tidier support for a number of new model objects and improves functionality of many existing tidiers!

New Tidiers

• Add tidiers for Rchoice objects (#961 by @vincentarelbundock and @Nateme16)
• Add tidiers for objects produced by car::leveneTest (#968 by @vincentarelbundock and @mkirzon)
• Add tidiers for objects produced by cmprsk::crr (#971 and #552 by @vincentarelbundock and @margarethannum)
• Add an augment() method for gam objects (#975 and #645 by @vincentarelbundock)
• Add tidiers for vars objects (#979 and #161 by @vincentarelbundock and @Diego-MX)

This release also restores tidiers for felm objects from the lfe package, which was recently unarchived from CRAN.

Improvements to existing tidiers

• tidy.emmGrid can now return std.error and conf.* columns at the same time. (#962 by @vincentarelbundock and @jmbarbone)
• tidy.garch can now produce confidence intervals (#964 by @vincentarelbundock and @IndrajeetPatil)
• tidy.coxph can now report confidence intervals on models utilizing penalized/clustering terms (#966 by @vincentarelbundock and @matthieu-faron)
• augment.lm now works when some regression weights are equal to zero (#965 by @vincentarelbundock and @vnijs)
• tidy.coxph can now handle models utilizing penalized/clustering terms (#966 and #969 by @vincentarelbundock, @matthieu-faron, and @KZARCA)
• Fix bug in tidy.speedglm on R 4.0.0+ (#974 by @uqzwang)
• tidy.multinom works with matrix response (#977 and #666 by @vincentarelbundock and @atyre2)
• Various bug fixes and improvements to documentation and errors.

broom 0.7.3

In broom 0.7.0, we introduced an error for model objects that subclassed lm and relied on tidy.lm(), or similarly for tidy.glm(). Tidiers for these objects were supported unintentionally, and we worried that tidiers for these objects would silently report innaccurate results.

In hindsight, this change was unnecessarily abrupt. We've decided to roll back this change, instead providing the following warning before allowing such objects to fall back to the lm/glm tidier methods:

Tidiers for objects of class {subclass} are not maintained by the broom team, and are only supported through the {dispatched method} tidier method. Please be cautious in interpreting and reporting broom output."

In addition,

• Restores tidiers for summary.lm objects (#953 by @grantmcdermott)
• Deprecate tidiers for the lfe package, which was archived from CRAN.
• Various bug fixes and improvements to documentation and errors.

broom 0.7.2

• In broom 0.7.0, we introduced an error for objects that subclassed
  lm and relied on tidy.lm(). We've transitioned this error to a warning.
• Various bug fixes and improvements to documentation and errors.

broom 0.7.1

While broom 0.7.1 is a minor release, it includes a number of exciting new
features and bug fixes!

New tidiers

• Add tidiers for margins objects. (#700 by @grantmcdermott)
• Added tidier methods for mlogit objects (#887 by @gregmacfarlane)
• Add glance.coeftest() method (#932 by @grantmcdermott)

Improvements to existing tidiers

One of the more major improvements in this release is the addition of the
interval argument to some augment methods for confidence, prediction,
and credible intervals. These columns will be consistently labeled .lower
and .upper! (#908 by @grantmcdermott, #925 by @bwiernik)

In addition...

• Extended the glance.aov() method to include an r.squared column!
• glance.survfit() now passes ... to summary.survfit() to allow for
  adjustment of RMST and other measures (#880 by @vincentarelbundock)
• Several unsupported model objects that subclass glm and lm now error
  more informatively.
• A number of improvements to documentation throughout the package.

Bug fixes

• Fixed newdata warning message in augment.*() output when the newdata
  didn't contain the response variable—augment methods no longer expect the
  response variable in the supplied newdata argument. (#897 by @rudeboybert)
• Fixed a bug related to tidy.geeglm() not being sensitive to the
  exponentiate argument (#867)
• Fixed augment.fixest() returning residuals in the .fitted column. The
  method also now takes a type.residuals argument and defaults to the same
  type.predict argument as the fixest predict() method. (#877 by @karldw)
• Fix tidy.felm confidence interval bug. Replaces "robust" argument with
  "se.type". (#919 by @grantmcdermott; supersedes #818 by @kuriwaki)
• Fix a bug in tidy.drc() where some term labels would result
  in the overwriting of entries in the curve column (#914)
• Fixed bug related to univariate zoo series in tidy.zoo() (#916 by @WillemVervoort)
• Fixed a bug related to tidy.prcomp() assigning the wrong PC labels from "loadings"
  and "scores" matrices (#910 by @tavareshugo)
• Fixed tidy.polr() bug where p-values could only be returned if
  exponentiate = FALSE.

Deprecations

We followed through with the planned deprecation of character vector tidiers
in this release. Other vector tidiers that were soft-deprecated in 0.7.0 will
be fully deprecated in a later release.