Added reconc_MixCond
, the implementation of Mixed conditioning for the reconciliation of mixed-type hierarchical forecasts.
Added reconc_TDcond
, the implementation of Top-down conditioning for the reconciliation of mixed-type hierarchical forecasts.
Vignette “Reconciliation of M5 hierarchy with mixed-type forecasts”.
Added functions PMF.get_mean
, PMF.get_quantile
, PMF.get_var
, PMF.sample
, PMF.summary
that return point estimates and samples from a probability mass function object.
Added the dataset M5_CA1_basefc
which contains the base forecasts for the store “CA1” of the hierarchical time series in the M5 competition.
Vignette “Properties of the reconciled distribution via conditioning”.
Added option in reconc_BUIS
to pass some base forecast as parameters and some as samples.
Added option in reconc_BUIS
to input a list of distributions instead of a string.
Fixed bugs and closed github issues.
Vignette “Probabilistic Reconciliation via Conditioning with bayesRecon”.
Added the schaferStrimmer_cov
function.
Fixed bugs.
Added a NEWS.md
file to track changes to the package.
Fixed issue that could cause test fails on some machines.