Receiver Operator Characteristics

ROC.factor

R Documentation

Description

The ROC()-function computes the tpr() and fpr() at thresholds provided by the \(response\)- or \(thresholds\)-vector. The function constructs a data.frame() grouped by \(k\)-classes where each class is treated as a binary classification problem.

Usage

## S3 method for class 'factor'
ROC(actual, response, thresholds = NULL, presorted = FALSE, ...)

## S3 method for class 'factor'
weighted.ROC(actual, response, w, thresholds = NULL, presorted = FALSE, ...)

## Generic S3 method
ROC(
 actual,
 response,
 thresholds = NULL,
 presorted  = FALSE,
 ...
)

## Generic S3 method
weighted.ROC(
 actual,
 response,
 w,
 thresholds = NULL,
 presorted  = FALSE,
 ...
)

Arguments

`actual`	A vector of values of length \(n\), and \(k\) levels.
`response`	A \(n \times k\) `<numeric>`-matrix. The estimated response probabilities for each class \(k\).
`thresholds`	An optional `<numeric>` vector of length \(n\) (default: NULL).
`presorted`	A -value length 1 (default: FALSE). If TRUE the input will not be sorted by threshold.
`…`	Arguments passed into other methods.
`w`	A `<numeric>`-vector of length \(n\). NULL by default.

Value

A data.frame on the following form,

`threshold`	`<numeric>` Thresholds used to determine `tpr()` and `fpr()`
`level`	The level of the actual
`label`	The levels of the actual
`fpr`	`<numeric>` The false positive rate
`tpr`	`<numeric>` The true positve rate

Examples

# 1) recode Iris
# to binary classification
# problem
iris$species_num <- as.numeric(
  iris$Species == "virginica"
)

# 2) fit the logistic
# regression
model <- glm(
  formula = species_num ~ Sepal.Length + Sepal.Width,
  data    = iris,
  family  = binomial(
    link = "logit"
  )
)

# 3) generate predicted
# classes
response <- predict(model, type = "response")

# 3.1) generate actual
# classes
actual <- factor(
  x = iris$species_num,
  levels = c(1,0),
  labels = c("Virginica", "Others")
)

# 4) generate reciever
# operator characteristics

# 4.1) calculate residual
# probability and store as matrix
response <- matrix(
  data = cbind(response, 1-response),
  nrow = length(actual)
)

# 4.2) construct 
# data.frame
roc <- ROC(
  actual   = actual,
  response = response
)

# 5) plot by species
plot(roc)

# 5.1) summarise
summary(roc)

# 6) provide custom
# threholds
roc <- ROC(
  actual     = actual,
  response   = response,
  thresholds = seq(
    1,
    0,
    length.out = 20
  )
)

# 5) plot by species
plot(roc)