\newcommand{\PR}{\Sexpr[results=rd]{tools:::Rd_expr_PR(#1)}}

\name{NEWS}
\title{broom.mixed News}
\encoding{UTF-8}

\section{CHANGES IN VERSION 0.2.9.4}{
  \itemize{
    \item minor changes only; test tweaks for CRAN compatibility
  }
}
\section{CHANGES IN VERSION 0.2.9.3}{
  \subsection{BUG FIXES}{
    \itemize{
      \item improved profile robustness in TMB tidiers
    }
  } % bug fixes
  \subsection{NEW FEATURES}{
    \itemize{
      \item \code{lme} tidier gets functionality for information about
      variance models (use \code{effects = "var_model"}) (Bill Denney)
      \item support for models with fixed sigma values in \code{lme}
      tidier (Bill Denney)
      \item added \code{tidy} and \code{glance} methods for
      \code{allFit} objects from the \code{lme4} package
      \item \code{get_methods()} function returns a table of all
      available tidy/glance/augment methods
    }
  } % new features
  \subsection{USER-VISIBLE CHANGES}{
    \itemize{
      \item improved \code{lme} tidying for random effects values
      \item \code{brms} tidiers no longer use deprecated
      \code{posterior_samples}
      \item {glance.lme4} now returns \code{nobs} (Cory Brunson)
      \item some tidiers are less permissive about unused arguments passed via \code{...}
    }
  } % user-visible changes
} % version 0.2.8

\section{CHANGES IN VERSION 0.2.7 (2021-07-07)}{
  \subsection{NEW FEATURES}{
    \itemize{
      \item experimental \code{TMB} tidiers (the \code{TMB} package
      does not return an object of class \code{TMB}, so users should run
      \code{class(fit) <- "TMB"} before attempting to tidy)
    }
  }
  \subsection{USER-VISIBLE CHANGES}{
    \itemize{
      \item term names are no longer "sanitized" in gamlss tidiers
      (e.g. "(Intercept)" is not converted to "X.Intercept.")
      \item gamlss \code{glance} method returns \code{nobs} (GH #113)
    }
  }
  \subsection{BUG FIXES}{
    \itemize{
      \item \code{lmerTest} Wald confidence intervals for \code{lmerTest}
      models now respect \code{ddf.method}
      \item \code{tidy.glmmTMB(.,effects="ran_vals")} fixed for
      \code{stringsAsFactors} changes in \code{glmmTMB} (GH #103)
      \item \code{tidy.gamlss} now works in a wider range of cases
      (GH #74)
      \item \code{tidy.brmsfit} works for models without group effects
      (GH #100)
    }
 } 
}
\section{CHANGES IN VERSION 0.2.6 (2020-05-17)}{
  \itemize{
    \item No improvements; compatibility with dplyr 1.0.0; skip examples
  }
}
\section{CHANGES IN VERSION 0.2.5 (2020-04-19)}{
  \subsection{NEW FEATURES}{
    \itemize{
      \item \code{lmer} tidier gets \code{ddf.method} (applies only to
      lmerTest fits)
      \item \code{glmmTMB} gets \code{exponentiate} options
      \item experimental \code{GLMMadaptive} tidiers
    }
  }
  \subsection{OTHER CHANGES}{
    \itemize{
      \item fixes for updates in tibble package
    }
  }
}
\section{CHANGES IN VERSION 0.2.4}{
  \subsection{NEW FEATURES}{
     \itemize{
	\item \code{gls} tidier gets confint (GH #49)
	}
      }
      \subsection{USER-VISIBLE CHANGES}{
	\itemize{
	  \item redundant \code{estimate.method} in MCMC tidiers goes
	  away; use \code{robust} to compute point estimates/uncertainty via
	  median and MAD rather than mean and SE
	}
      }
      \subsection{BUG FIXES}{
	\itemize{
          \item misc fixes: \code{lme4} tidiers (confint for ran_vals,
	  profile conf intervals fixed), \code{R2jags}, \code{gamlss} ...
	  \item ran_vals works for \code{glmmTMB}
	}
      }
}
\section{CHANGES IN VERSION 0.2.3}{
  \subsection{BUG FIXES}{
    \itemize{
      \item don't ignore conf.level in tidy.(merMod|glmmTMB) (GH #30,31: @strengejacke)
      \item levels correct in tidy.brmsfit (GH #36: @strengejacke)
      \item component argument works for random effects in glmmTMB (GH #33: @strengejacke)
    }
  }
  \subsection{NEW FEATURES}{
    \itemize{
      \item \code{brmsfit} and \code{rstanarm} methods allow
      \code{conf.method="HPDinterval"}
    }
  }
  \subsection{USER-VISIBLE CHANGES}{
    \itemize{
      \item tidy.brmsfit gets component column (GH #35: @strengejacke),
      response column for multi-response models (GH #34: @strengejacke)
      \item component tags are stripped from tidied brmsfit objects
      \item "Intercept" terms in brms fits are re-coded as
      "(Intercept)" by default, for dotwhisker/cross-model
      compatibility; for previous behaviour, specify \code{fix.intercept=FALSE}
    }
  }
}
\section{CHANGES IN VERSION 0.2.2}{
  \itemize{
    \item modify examples, for CRAN compliance
  }
}
\section{CHANGES IN VERSION 0.2.1}{
  \itemize{
    \item reduced size of stored fits for examples, for CRAN compliance
  }
}
\section{CHANGES IN VERSION 0.2.0}{
  \subsection{NEW FEATURES}{
    \itemize{
      \item more consistent term names in \code{brmsfit}, \code{rstanreg} tidiers
      \item improved \code{tidy.MCMCglmm}
    }
  }
  \subsection{USER-VISIBLE CHANGES}{
    \itemize{
      \item all methods return tibbles (\code{tbl_df}) rather than data
      frames
      \item the value of the \code{group} variable for fixed-effect
      parameters has changed from \code{"fixed"} to \code{NA}
      \item \code{brmsfit} and \code{rstanarm} tidiers are more
      consistent with other tidiers (e.g. \code{conf.level} rather than
      \code{prob})
    }
  }
}
\section{CHANGES IN VERSION 0.0.1}{
  \subsection{BUG FIXES}{
    \itemize{
      \item Sorted out some of the confusion over random effect naming:
      "ran_vals" extracts conditional modes/BLUPs/varying parameters
      (deviations from population-level estimates),
      while "ran_coefs" extracts group-level estimates
    }
  }
  \subsection{NEW FEATURES}{
     \itemize{
       \item improved \code{nlme} tidiers
       \item improved \code{glmmTMB} tidiers (can handle some
       zero-inflation parameters)
       \item \code{lme4} tidiers now optionally take a pre-computed
       \code{profile} argument when using \code{conf.method="profile"}

     }
  }
  \subsection{USER-VISIBLE CHANGES}{
    \itemize{
      \item The default behaviour of most mixed-model tidiers has
      changed/will gradually be changed to the following
      (description modified from TJ Mahr at
      \url{https://github.com/tidymodels/broom/issues/96}):
      \itemize{
	\item Random effect variances and covariances can now be extracted. 
	\code{effects = "ran_pars"} returns the standard deviations/correlations
	of random effects (if \code{scales="sdcor"} [default])
	or their variances
	and covariances (if \code{scales = "varcov"})
	\item Random effects estimates are now extracted with 
	\code{effects = "ran_coefs"} for the group-level
	estimates (previously these effects were 
	extracted with \code{tidy(model, "random")}) or
	\code{effects = "ran_vals"} for the conditional modes
	(deviations of the group-level parameters from the
	population-level estimates)
	\item \code{effects} can take a vector of values (those listed
	above, plus "fixed" for fixed effects). The default value
	is \code{effects = c("ran_pars", "fixed")} which extracts random 
	effect variances/covariances and fixed effect estimates.
      }
      \item \code{term} names for random-effect parameters no longer
      contain a (redundant) group specifier (at least for \code{lme4}
      models); use something like
      \code{tidyr::unite(term,term,group,sep=".")} 
      to collapse the two columns
    }
  }
}