The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
mat and data are provided,
mat is used for network estimation, while data
is only used for additional calculations (e.g., for compatibility with
bootnet).cor_calc() now automatically scales raw data when no
variables are treated as ordered. This was already the case when called
indirectly, but now also applies when calling it directly. Only affects
the returned means, not the estimated correlations.means argument to
regularization_net(). Not intended for typical use, but
needed for integration with bootnet.ns no longer accepts vectors in
regularization_net(), instead accepts a matrix.network_vars and auxiliary_vars
arguments to cor_calc(), neighborhood_net(),
and regularization_net(). These arguments accept both
variable names and numeric indices referring to the provided data. ##
New featuresmids object when
using stacked multiple imputation.network_vars and
auxiliary_vars arguments, users can now specify which
variables are used for network estimation and which additional variables
are included for correlation estimation in the presence of missing
data.k, which controlled the penalty term in
information-criterion calculations, has been removed for security
reasons. Instead, the penalty type is now specified via the argument
ic_type (see the corresponding help pages).ridge penalty in
the multiple-imputation pmm workflow when looking for
donors through regressions (mice). Instead of forcing
ridge = 0, the function now uses the default value defined
by mice, ensuring consistent and method-appropriate
regularization.reg_network() for network estimation
using regularization, supporting both convex and non-convex penalties as
well as multiple options for computing the likelihood in the information
criterion when missing values are present.ordered_suggest(), a heuristic procedure
for identifying variables that may be treated as ordered categorical
based on their distribution and available information.mantar_dummy_full_cat and
mantar_dummy_mis_cat, containing only ordered categorical
variables (with and without missing values).mantar_dummy_full_mix and
mantar_dummy_mis_mix, containing mixtures of ordered
categorical and continuous variables (with and without missing
values).mantar_dummy_full_cont and
mantar_dummy_mis_cont to better reflect that they contain
only continuous variables.These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.