WGCNA (version 1.61)

accuracyMeasures: Accuracy measures for a 2x2 confusion matrix or for vectors of predicted and observed values.

Description

The function calculates various prediction accuracy statistics for predictions of binary or quantitative (continuous) responses. For binary classification, the function calculates the error rate, accuracy, sensitivity, specificity, positive predictive value, and other accuracy measures. For quantitative prediction, the function calculates correlation, R-squared, error measures, and the C-index.

Usage

accuracyMeasures(
  predicted, 
  observed = NULL, 
  type = c("auto", "binary", "quantitative"),
  levels = if (isTRUE(all.equal(dim(predicted), c(2,2)))) colnames(predicted)
            else if (is.factor(predicted))
              sort(unique(c(as.character(predicted), as.character(observed))))
            else sort(unique(c(observed, predicted))),
  negativeLevel = levels[2], 
  positiveLevel = levels[1] )

Arguments

predicted

either a a 2x2 confusion matrix (table) whose entries contain non-negative integers, or a vector of predicted values. Predicted values can be binary or quantitative (see type below). If a 2x2 matrix is given, it must have valid column and row names that specify the levels of the predicted and observed variables whose counts the matrix is giving (e.g., the function table sets the names appropriately.) If it is a 2x2 table and the table contains non-negative real (non-integer) numbers the function outputs a warning.

observed

if predicted is a vector of predicted values, this (observed) must be a vector of the same length giving the "gold standard" (or observed) values. Ignored if predicted is a 2x2 table.

type

character string specifying the type of the prediction problem (i.e., values in the predicted and observed vectors). The default "auto" decides type automatically: if predicted is a 2x2 table or if the number of unique values in the concatenation of predicted and observed is 2, the prediction problem (type) is assumed to be binary, otherwise it is assumed to be quantitative. Inconsistent specification (for example, when predicted is a 2x2 matrix and type is "quantitative") trigger errors.

levels

a 2-element vector specifying the two levels of binary variables. Only used if type is "binary" (or "auto" that results in the binary type). Defaults to either the column names of the confusion matrix (if the matrix is specified) or to the sorted unique values of observed and opredicted.

negativeLevel

the binary value (level) that corresponds to the negative outcome. Note that the default is the second of the sorted levels (for example, if levels are 1,2, the default negative level is 2). Only used if type is "binary" (or "auto" that results in the binary type).

positiveLevel

the binary value (level) that corresponds to the positive outcome. Note that the default is the second of the sorted levels (for example, if levels are 1,2, the default negative level is 2). Only used if type is "binary" (or "auto" that results in the binary type).

Value

Data frame with two columns:

Measure

this column contais character strings that specify name of the accuracy measure.

Value

this column contains the numeric estimates of the corresponding accuracy measures.

Details

The rows of the 2x2 table tab must correspond to a test (or predicted) outcome and the columns to a true outcome ("gold standard"). A table that relates a predicted outcome to a true test outcome is also known as confusion matrix. Warning: To correctly calculate sensitivity and specificity, the positive and negative outcome must be properly specified so they can be matched to the appropriate rows and columns in the confusion table.

Interchanging the negative and positive levels swaps the estimates of the sensitivity and specificity but has no effect on the error rate or accuracy. Specifically, denote by pos the index of the positive level in the confusion table, and by neg th eindex of the negative level in the confusion table. The function then defines number of true positives=TP=tab[pos, pos], no.false positives =FP=tab[pos, neg], no.false negatives=FN=tab[neg, pos], no.true negatives=TN=tab[neg, neg]. Then Specificity= TN/(FP+TN) Sensitivity= TP/(TP+FN) NegativePredictiveValue= TN/(FN + TN) PositivePredictiveValue= TP/(TP + FP) FalsePositiveRate = 1-Specificity FalseNegativeRate = 1-Sensitivity Power = Sensitivity LikelihoodRatioPositive = Sensitivity / (1-Specificity) LikelihoodRatioNegative = (1-Sensitivity)/Specificity. The naive error rate is the error rate of a constant (naive) predictor that assigns the same outcome to all samples. The prediction of the naive predictor equals the most frequenly observed outcome. Example: Assume you want to predict disease status and 70 percent of the observed samples have the disease. Then the naive predictor has an error rate of 30 percent (since it only misclassifies 30 percent of the healthy individuals).

References

http://en.wikipedia.org/wiki/Sensitivity_and_specificity

Examples

# NOT RUN {
m=100
trueOutcome=sample( c(1,2),m,replace=TRUE)
predictedOutcome=trueOutcome
# now we noise half of the entries of the predicted outcome
predictedOutcome[ 1:(m/2)] =sample(predictedOutcome[ 1:(m/2)] )
tab=table(predictedOutcome, trueOutcome) 
accuracyMeasures(tab)

# Same result:
accuracyMeasures(predictedOutcome, trueOutcome)

# }