Learn R Programming

klaR (version 1.7-3)

woe: Weights of evidence

Description

Computes weight of evidence transform of factor variables for binary classification.

Usage

woe(x, ...)
# S3 method for default
woe(x, grouping, weights = NULL, zeroadj = 0, ids = NULL, 
                      appont = TRUE, ...)
# S3 method for formula
woe(formula, data = NULL, weights = NULL, ...)

Value

Returns an object of class woe that can be applied to new data.

woe

WOE coefficients for factor2numeric transformation of each (specified) variable.

IV

Vector of information values of all transformed variables.

newx

Data frame of transformed data if appont.

Arguments

x

A matrix or data frame containing the explanatory variables.

grouping

A factor specifying the binary class for each observation.

formula

A formula of the form grouping ~ x1 + x2 + ... That is, the response is the grouping factor and the right hand side specifies the discriminators.

data

Data frame from which variables specified in formula are to be taken.

weights

Vector with observation weights. For call woe(formula, data, weights) also a a character is possible that specifies the name of the weight coloumn in data.

zeroadj

Additive constant to be added for a level with 0 observations in a class.

ids

Vector of either indices or variable names that specifies the variables to be transformed.

appont

Application on training data: logical indicating whether the transformed values for the training data should be returned by recursive calling of predict.woe.

...

For woe.formula: Further arguments passed to function woe.default such as ids. For woe.default: replace = FALSE can be passed to recursive call of predict.woe if appont.

Author

Gero Szepannek

Details

To each factor level \(x\) a numeric value \(WOE(x) = ln(f(x|1)/f(x|2))\) is assigned where 1 and 2 denote the class labels. The WOE transform is motivated for subsequent modelling by logistic regression. Note that the frequencies of the classes should be investigated before. Information values heuristically quantify the discriminatory power of a variable by \(IV = (f(x|1)-f(x|2)) ln(f(x|1)/f(x|2))\).

References

Good, I. (1950): Probability and the Weighting of Evidences. Charles Griffin, London.

Kullback, S. (1959): Information Theory and Statistics. Wiley, New York.

See Also

predict.woe, plot.woe

Examples

Run this code
## load German credit data
data("GermanCredit")

## training/validation split
train <- sample(nrow(GermanCredit), round(0.6*nrow(GermanCredit)))

woemodel <- woe(credit_risk~., data = GermanCredit[train,], zeroadj=0.5, applyontrain = TRUE)
woemodel

## plot variable information values and woes
plot(woemodel)
plot(woemodel, type = "woes")

## apply woes 
traindata <- predict(woemodel, GermanCredit[train,], replace = TRUE)
str(traindata)

## fit logistic regression model
glmodel     <- glm(credit_risk~., traindata, family=binomial)
summary(glmodel)
pred.trn <- predict(glmodel, traindata, type = "response")

## predict validation data
validata <- predict(woemodel, GermanCredit[-train,], replace = TRUE)
pred.val <- predict(glmodel, validata, type = "response")

Run the code above in your browser using DataLab