diagnostic odds ratio

dor.factor

R Documentation

Description

The dor()-function computes the Diagnostic Odds Ratio (DOR), a single indicator of test performance, between two vectors of predicted and observed factor() values. The weighted.dor() function computes the weighted diagnostic odds ratio.

When aggregate = TRUE, the function returns the micro-average DOR across all classes \(k\). By default, it returns the class-wise DOR.

Usage

## S3 method for class 'factor'
dor(actual, predicted, ...)

## S3 method for class 'factor'
weighted.dor(actual, predicted, w, ...)

## S3 method for class 'cmatrix'
dor(x, ...)

dor(...)

weighted.dor(...)

Arguments

actual	A vector of <factor>- of length \(n\), and \(k\) levels.
predicted	A vector of <factor>-vector of length \(n\), and \(k\) levels.
…	Arguments passed into other methods
w	A <numeric>-vector of length \(n\). NULL by default.
x	A confusion matrix created cmatrix().

Value

If micro is NULL (the default), a named <numeric>-vector of length k

If micro is TRUE or FALSE, a <numeric>-vector of length 1

Calculation

The metric is calculated for each class \(k\) as follows,

\[ \text{DOR}_k = \frac{\text{PLR}_k}{\text{NLR}_k} \]

Where \(\text{PLR}_k\) and \(\text{NLR}_k\) is the positive and negative likelihood ratio for class \(k\), respectively. See plr() and nlr() for more details.

When aggregate = TRUE, the micro-average is calculated as,

\[ \overline{\text{DOR}} = \frac{\overline{\text{PLR}_k}}{\overline{\text{NLR}_k}} \]

Where \(\overline{\text{PLR}}\) and \(\overline{\text{NLR}}\) is the micro-averaged is the positive and negative likelihood ratio, respectively.

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
predicted <- factor(
  as.numeric(
    predict(model, type = "response") >` 0.5
  ),
  levels = c(1,0),
  labels = c("Virginica", "Others")
)

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


# 4) evaluate model performance
# with Diagnostic Odds Ratio
cat("Diagnostic Odds Ratio", sep = "\n")
dor(
  actual    = actual, 
  predicted = predicted
)

cat("Diagnostic Odds Ratio (weighted)", sep = "\n")
weighted.dor(
  actual    = actual,
  predicted = predicted,
  w         = iris$Petal.Length/mean(iris$Petal.Length)
)