averaging_checks: Check if quantile forecasts are commensurate across groups

In dshemetov/evalcast-mirror: Tools For Evaluating COVID Forecasters

Description

Check if quantile forecasts are commensurate across groups

Usage

averaging_checks(
  cards,
  grp_vars = c("forecaster", "forecast_date", "ahead"),
  avg_vars = c("geo_value"),
  return_checks = FALSE
)

Arguments

cards	a data frame containing at least the columns named in grp_vars, avg_vars and a column called quantile
grp_vars	character vector of named columns in the score_card at which average performance will be returned
avg_vars	character vector of named columns in the score_card over which averages will be computed
return_checks	useful for debugging. returns a data frame indicating which groups have valid or invalid avg_vars

Value

filtered cards based on which groups are completely valid (meaning "correct" quantiles are present for all avg_vars in the group)

dshemetov/evalcast-mirror documentation built on Feb. 4, 2022, 8:52 a.m.