From 8bd0d94c30c1f9b0cd6d580fea072f75a26f4742 Mon Sep 17 00:00:00 2001 From: "Pavel N. Krivitsky" Date: Mon, 23 Dec 2024 10:59:09 +1100 Subject: [PATCH] Updated NEWS. --- inst/NEWS.Rd | 110 +++++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 98 insertions(+), 12 deletions(-) diff --git a/inst/NEWS.Rd b/inst/NEWS.Rd index cec235c7..e43bb64f 100644 --- a/inst/NEWS.Rd +++ b/inst/NEWS.Rd @@ -11,6 +11,15 @@ \title{NEWS file for the \code{\link[=ergm-package]{ergm}} package} \encoding{UTF-8} + +%% Note: This needs to be kept in sync with man/macros/ergmTerm.Rd +\newcommand{\ergmTerm}{\code{\link[#1:#2-ergmTerm]{#2#3}}} +\newcommand{\ergmConstraint}{\code{\link[#1:#2-ergmConstraint]{#2#3}}} +\newcommand{\ergmReference}{\code{\link[#1:#2-ergmReference]{#2#3}}} +\newcommand{\ergmHint}{\code{\link[#1:#2-ergmHint]{#2#3}}} +\newcommand{\ergmProposal}{\code{\link[#1:#2-ergmProposal]{#2#3}}} + + % When a release is forked, copy the sections to a new release and % delete unused sections. @@ -65,6 +74,83 @@ %% } +\section{Changes in version 4.8.0}{ + + \subsection{NEW FEATURES}{ + \itemize{ + \item{ + A new bipartite term operator \ergmTerm{ergm}{Project}{()}, with aliases \ergmTerm{ergm}{Proj1}{()} and \ergmTerm{ergm}{Proj2}{()}, evaluating the statistics on an undirected valued network of counts constructed by projecting the bipartite network onto one of its modes. + } + \item{ + A new family of terms, \ergmTerm{ergm}{nodecovrange}{()}, \ergmTerm{ergm}{nodeocovrange}{()}, \ergmTerm{ergm}{nodeicovrange}{()}, \ergmTerm{ergm}{b1covrange}{()}, and \ergmTerm{ergm}{b2covrange}{()}, to evaluate the sum over the nodes of the range of the specified quantitative attribute value among the node's neighbors. (Thanks to Marion Hoffman for the idea!) + } + \item{ + A new family of terms, \ergmTerm{ergm}{nodefactordistinct}{()}, \ergmTerm{ergm}{nodeofactordistinct}{()}, \ergmTerm{ergm}{nodeifactordistinct}{()}, \ergmTerm{ergm}{b1factordistinct}{()}, and \ergmTerm{ergm}{b2factordistinct}{()}, to evaluate the sum over the nodes of the number of distinct values of a categorical factor among the node's neighbors. (Thanks to Marion Hoffman for the idea!) + } + \item{ + The \ergmConstraint{ergm}{fixallbut}{()} now supports an \code{\link{rlebdm}} (in addition to an edge list matrix and a \code{\link[network]{network}}) for its argument. + } + } + } + + + \subsection{BUG FIXES}{ + \itemize{ + \item{ + Fix in stochastic approximation code for compatibility with most recent \CRANpkg{coda}. + } + \item{ + In \code{\link[ergm]{ergm()}}, a bug in initialization of MH proposals for constrained sampling requesting auxiliaries has been fixed. + } + } + } + + + \subsection{OTHER USER-VISIBLE CHANGES}{ + \itemize{ + \item{ + Precision of geometrically weighted statistics (degrees and shared partners) has been improved, particularly for large decay parameters. + } + \item{ + Dynamic term documentation is now more robust to extension developer error. + } + \item{ + A new \code{\link[=as.rle.rlebdm]{as.rle()}} method for \code{\link{rlebdm}} objects to cast it back to an \code{\link{rle}}. + } + } + } + + \subsection{C-LEVEL FACILITIES}{ + \itemize{ + \item{ + Degree-conditioned proposals are now exported in \file{inst/include/}. + } + \item{ + The M-H proposal API now makes lengths of input vectors available to the proposals, paralleling the change statistics API. + } + } + } + + %% \subsection{UTILITIES}{ + %% \itemize{ + %% \item Likewise, "UTILITIES" is probably unnecessary. + %% } + %% } + + %% \subsection{INSTALLATION}{ + %% \itemize{ + %% \item ... as is "INSTALLATION". + %% } + %% } + + %% \subsection{PACKAGE INSTALLATION}{ + %% \itemize{ + %% \item ditto. + %% } + %% } + +} + \section{Changes in version 4.7.5}{ @@ -105,7 +191,7 @@ \code{\link[=blockdiag-ergmConstraint]{blockdiag}} constraint initialization now detects when blocks are non-contiguous and stops with an error. (Previously, it would behave in undefined ways.) } \item{ - \code{\link[=Sum-ergmTerm]{Sum}} operator now handles \code{\link{I}()} label specifications correctly; this also fixes an error in the \code{\link[=Prod-ergmTerm]{Prod}} term. + \ergmTerm{ergm}{Sum}{} operator now handles \code{\link{I}()} label specifications correctly; this also fixes an error in the \ergmTerm{ergm}{Prod}{} term. } \item{ Detection of dyadic dependence now ignores dyadic dependence of hints (since they do not affect the sample space) and auxiliaries (since they do not, in and of themselves, add dyad-dependent statistics). @@ -123,7 +209,7 @@ In proposal help, listing of proposal table entries works again. } \item{ - \code{\link[=nodemix-ergmTerm]{nodemix}()} now uses correct parameter names. + \ergmTerm{ergm}{nodemix}{()} now uses correct parameter names. } \item{ Godambe information for the MPLE now works if the model has an offset. (Thanks, Cornelius Fritz for reporting and Michał Bojanowski for fixing.) @@ -132,7 +218,7 @@ When constructing the starting networks for missing data MCMC, imputation of dyads is skipped if the constraints are dyad-dependent, since inserting edges may break the constraint. } \item{ - \code{\link[=edgecov-ergmTerm]{edgecov}()} and \code{\link[=dyadcov-ergmTerm]{dyadcov}()} now handle \code{\link[network]{network}}-format input correctly. + \ergmTerm{ergm}{edgecov}{()} and \ergmTerm{ergm}{dyadcov}{()} now handle \code{\link[network]{network}}-format input correctly. } \item{ Missing data MLE code can now handle the scenarios in which the statistics in the constrained sample are constant and/or the statistics in the unconstrained sample are highly correlated. @@ -144,10 +230,10 @@ Likelihood calculation is now robust to dropped parameters and parameters fixed at infinity. } \item{ - \code{\link[Label-ergmTerm]{Label}()} operator now documents its behavior when the model is curved more clearly. + \ergmTerm{ergm}{Label}{()} operator now documents its behavior when the model is curved more clearly. } \item{ - Invalid \code{levels2} specification for \code{\link[=mm-ergmTerm]{mm()}} no longer causes memory errors. + Invalid \code{levels2} specification for \ergmTerm{ergm}{mm}{()} no longer causes memory errors. } \item{ Valued proposal updater function was not being passed the current edge state. @@ -167,7 +253,7 @@ \code{\link{LARGEST}}, \code{\link{SMALLEST}}, and \code{\link{COLLAPSE_SMALLEST}} now break ties lexicographically with a warning. } \item{ - \code{\link[=absdiffcat-ergmTerm]{absdiffcat}} is now more memory-efficient during initialization. + \ergmTerm{ergm}{absdiffcat}{} is now more memory-efficient during initialization. } \item{ MCMLE estimation code for missing data MLE with high missingness fraction is more robust. @@ -191,7 +277,7 @@ \code{\link{ergm_Init_stop}()}, \code{\link{ergm_Init_warning}()}, and \code{\link{ergm_Init_message}()} that behave more like their \pkg{base} counterparts have been added to the API. \code{\link{ergm_Init_abort}()}, \code{\link{ergm_Init_warn}()}, and \code{\link{ergm_Init_info}()} will eventually change to use their \CRANpkg{rlang} semantics. } \item{ - \code{\link[=edgecov-ergmTerm]{edgecov}()} and \code{\link[=dyadcov-ergmTerm]{dyadcov}()} now check that their covariate matrix has the correct dimension. + \ergmTerm{ergm}{edgecov}{()} and \ergmTerm{ergm}{dyadcov}{()} now check that their covariate matrix has the correct dimension. } \item{ Constraints \code{\link[fixedas-ergmConstraint]{fixedas}()} and \code{\link[fixedas-ergmConstraint]{fixallbut}()} now warn when given a network whose size does not match the LHS network's. @@ -493,7 +579,7 @@ New functions \code{\link{search.ergmHints}()} and \code{\link{search.ergmReferences}()} can now be used to search available \code{\link{ergmHints}} and \code{\link{ergmReferences}}, respectively. } \item{ - A new \code{\link[=For-ergmTerm]{For}()} term operator to construct a list of terms with some varying parameter. + A new \ergmTerm{ergm}{For}{()} term operator to construct a list of terms with some varying parameter. } \item{ Term API vignette is now more complete, and includes all API elements, not just the post-4.0 ones. @@ -594,7 +680,7 @@ \subsection{NEW FEATURES}{ \itemize{ \item{ - In terms \code{\link[=edgecov-ergmTerm]{edgecov}()} and \code{\link[=dyadcov-ergmTerm]{dyadcov}()}, the argument can now be a network. + In terms \ergmTerm{ergm}{edgecov}{()} and \ergmTerm{ergm}{dyadcov}{()}, the argument can now be a network. } } } @@ -704,7 +790,7 @@ } \item{ - Term \code{\link[=nodemix-ergmTerm]{nodemix}} can now be passed a \code{levels2} argument that is a \code{\link{factor}} or \code{\link{character}} and optionally a matrix, allowing multiple cells to be mapped to the same statistic. (Joyce Cheng) + Term \ergmTerm{ergm}{nodemix}{} can now be passed a \code{levels2} argument that is a \code{\link{factor}} or \code{\link{character}} and optionally a matrix, allowing multiple cells to be mapped to the same statistic. (Joyce Cheng) } \item{ @@ -733,7 +819,7 @@ } \item{ - \code{\link[=edgecov-ergmTerm]{edgecov}} and \code{\link[=dyadcov-ergmTerm]{dyadcov}} now detect and stop with an error when the specified network attribute is not found. + \ergmTerm{ergm}{edgecov}{} and \ergmTerm{ergm}{dyadcov}{} now detect and stop with an error when the specified network attribute is not found. } \item{ @@ -813,7 +899,7 @@ } \item{ - \code{\link{check.ErgmTerm}()} helper function can now also capture the expressions resulting in term arguments. This incidentally fixes a regression in statistic naming of \code{\link{edgecov-ergmTerm}} and \code{\link{dyadcov-ergmTerm}} effects. + \code{\link{check.ErgmTerm}()} helper function can now also capture the expressions resulting in term arguments. This incidentally fixes a regression in statistic naming of \ergmTerm{ergm}{edgecov}{()} and \ergmTerm{ergm}{dyadcov}{()} effects. } \item{