cat.slice: Slice categorical variable

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

Slice categorical variable

Description

cat.slice implements manual re-coding of character vector values for a given mapping scheme. This procedure is one of the helper functions which are handy for the model monitoring phase (i.e. after model implementation).

Usage

cat.slice(x, mapping, sc = NA, sc.r = "SC")

Arguments

x	Character vector to be re-coded.
mapping	Data frame with compulsory columns: x.orig and x.mapp which represent the mapping scheme. Column x.orig should contain unique values of original vector x, while x.mapp should contain corresponding mapping values.
sc	Character vector with special case elements. Default value is NA.
sc.r	Character vector used for replacement of special cases. If supplied as one element vector, it will be recycled to the length of sc. Default value is "SC".

Value

The command cat.slice returns vector of re-coded values and special cases.

Examples

suppressMessages(library(PDtoolkit))
data(gcd)
x <- gcd$maturity
#artificially add some special values
x[1:5] <- Inf
x[6:7] <- NA
mbin <- cum.bin(x = x, y = gcd$qual, sc.method = "together")
mbin[[1]]
gcd$x <- mbin[[2]]
cb <- cat.bin(x = gcd$x, 
	  y = gcd$qual, 
	  sc = "SC",
	  sc.merge = "none",
	  min.pct.obs = 0.05, 
	  min.avg.rate = 0.05)
x <- gcd$x
mapping <- data.frame(x.orig = x, x.mapp = cb[[2]])%>%
       group_by(x.orig, x.mapp) %>%
       summarise(n = n(), .groups = "drop")
mapping <- data.frame(mapping[, -3])
sc <- cat.slice(x = x, 
	     mapping = mapping, 
	     sc = NA, 
	     sc.r = "SC")
#compare automatic and manual re-coding
table(cb[[2]], useNA = "always")
table(sc, useNA = "always")

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