ratio2cperc: Calculate C.perc from ratio

In mrfoliveira/STResampling-JDSA2020: Biased Resampling Strategies for Imbalanced Spatio-Temporal Forecasting

Description

Calculate C.perc from ratio

Usage

ratio2cperc(ratio, y.relev, rel.thr, type)

Arguments

ratio	Value that defines the target ratio between the number of extreme values and the number of normal values. If more than one bump exists: re-sampling will be done proportionally to each bump, so that the ratio between the sum of all extreme values and all normal values in the final data set is the target ratio.
y.relev	A vector containing the relevance values of target
rel.thr	A relevance threshold
type	"under", "over", or "gauss"

Value

A list with C.perc, the re-sampling percentage that will result in the target ratio and orig_ratio indicating the original ratio between extreme and normal values

mrfoliveira/STResampling-JDSA2020 documentation built on June 28, 2021, 7:01 p.m.