ush.bin: U-shape binning algorithm

In PDtoolkit: Collection of Tools for PD Rating Model Development and Validation

U-shape binning algorithm

Description

ush.bin performs U-shape binning. All algorithms from monobin package are available. Due to specific nature of binning algorithms it is possible that for some selected knots algorithm will not be able to find U-shape. Therefore, users are encourage to inspect the results more into details and to try different binning algorithms.

Usage

ush.bin(
  x,
  y,
  knot,
  method,
  sc = c(NA, Inf, -Inf, NaN),
  sc.method = "together",
  g = 20,
  min.pct.obs = 0.05,
  min.avg.rate = 0.01,
  p.val = 0.05,
  woe.trend = TRUE,
  woe.gap = 0.1
)

Arguments

x	Numeric vector to be binned.
y	Numeric target vector (binary).
knot	Numeric value of selected knot. Usually the results of ush.test function.
method	Binning method. Available options are all from monobin package: "cum.bin", "iso.bin", "ndr.bin", "pct.bin", "sts.bin", "woe.bin", "mdt.bin".
sc	Numeric vector with special case elements. Default values are c(NA, NaN, Inf, -Inf). Recommendation is to keep the default values always and add new ones if needed. Otherwise, if these values exist in x and are not defined in the sc vector, function can report the error.
sc.method	Define how special cases will be treated, all together or separately. Possible values are "together", "separately".
g	Number of starting groups. Only needed for "cum.bin", "pct.bin" and mdt.bin methods. Default is 20.
min.pct.obs	Minimum percentage of observations per bin. Default is 0.05 or 30 observations.
min.avg.rate	Minimum y average rate. Default is 0.05 or 30 observations.
p.val	Threshold for p-value. Only needed for "sts.bin" and "ndr.bin" methods. Default is 0.05.
woe.trend	Logical. Only needed for "pct.bin" method with default TRUE.
woe.gap	Minimum WoE gap between bins. Only needed for "woe.bin" method with default of 0.1.

Value

The command ush.bin generates a list of two objects. The first object, data frame summary.tbl presents a summary table of final binning, while x.trans is a vector of discretized values.

Examples

res <- ush.bin(x = gcd$amount, y = gcd$qual, knot = 2992.579, method = "ndr.bin")
res[[1]]
plot(res[[1]]$dr, type = "l")

PDtoolkit documentation built on Sept. 20, 2023, 9:06 a.m.