table_compare: Construct a 2x2 contingency table and calculate associated...
In tpepler/theolib: Collection of Auxiliary Functions

Description Usage Arguments Value Note Author(s) Examples

View source: R/table_compare.R

Construct a contingency table to compare predicted outcomes to actual/observed outcomes. Also calculates statistics associated with the table.

1	table_compare(pred, actual)

`pred`	Predicted outcomes (0/1), as obtained from a model or decision rule.
`actual`	Actual observed outcomes (0/1).

Returns a list with the following:

`cont.table`	The 2x2 contingency table
`sens`	Sensitivity
`spec`	Specificity
`misclass.rate`	Misclassification rate
`false.pos`	False positive rate
`false.neg`	False negative rate
`ppv`	Positive predictive value
`npv`	Negative predictive value

Note that the calculations assume that a "positive" result is coded as a 1, and a "negative" result as a 0.

Theo Pepler

1
2
3

predict.outcomes <- c(1, 1, 0, 1, 1, 1, 0, 0)
actual.outcomes <- c(1, 1, 1, 1, 0, 0, 0, 0)
table_compare(pred = predict.outcomes, actual = actual.outcomes)