Package index • alloscore2

All functions

add_pdqr_funs(): Add list columns of p/d/q/r functions to a forecast data frame
allocate(): Allocate a budget to minimize expected gpl loss
allocate_model_out(): Allocate a budget across hubverse model output
alloscore(): Score allocations against realized outcomes
alloscore_model_out(): Score hubverse model output with an allocation scoring rule
as_alloscore_df(): Convert hubverse model output into forecasts alloscore2 can allocate against
dexp_gpl_df(): Create a list of derivatives of expected gpl losses from a gpl_df
dexp_gpl_loss(): Derivative of the expected gpl loss
dexp_over_loss(): Derivative of the expected over-prediction loss
dexp_under_loss(): Derivative of the expected under-prediction loss
exp_gpl_loss_fun(): Create an expected gpl loss function
exp_over_loss(): Expected g-linear loss for over-prediction of a random outcome
exp_under_loss(): Expected g-linear loss for under-prediction of a random outcome
gpl(): Get the gpl loss data frame of an allocation
gpl_loss_fun(): Create a generalized piecewise linear (gpl) loss function
meb_gpl_df(): Create marginal expected benefit functions for a gpl_df
new_gpl_df(): Create a data frame of gpl loss functions and their parameters
oracle_allocate(): Allocate as an oracle that knows the observed outcomes
oracle_allocate_direct(): Find the oracle allocation directly
oracle_alloscore_direct(): Find and score the oracle allocation directly
over_loss(): Basic g-linear loss for over-prediction of an outcome
pdqr_factory(): Build a p/d/q/r function with parameters fixed
plot_components(): Plot the per-target components of an allocation score
plot_components_slim(): Plot the components of a slim scored allocation
plot_iterations(): Plot the allocation search for one budget
plot_scores_slim(): Plot allocation scores against the budget or over time
post_process(): Repair allocations left infeasible by plateaus in the objective
slim(): Drop the heavy list columns of an allocated data frame
stdize_met_params(): Convert meteorologist parameters to kappa and alpha
stdize_news_params(): Convert newsvendor parameters to kappa and alpha
stdize_ou_params(): Convert over/under-prediction costs to kappa and alpha
under_loss(): Basic g-linear loss for under-prediction of an outcome
weights(<allocated>): Get the weights of an allocated data frame
zxh_tab2: Multiproduct newsboy data with Normal demand
zxh_tab3: Multiproduct newsboy data with Beta demand