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.
Need a high-speed mirror for your open-source project?
Contact our mirror admin team at info@clientvps.com.
This archive is provided as a free public service to the community.
Proudly supported by infrastructure from VPSPulse , RxServers , BuyNumber , UnitVPS , OffshoreName and secure payment technology by ArionPay.