misclass {simex} | R Documentation |
Takes a data.frame
and produces misclassified data.
Probabilities for the missclassification are given in the mc.matrix
.
misclass(data.org, mc.matrix, k)
data.org |
data.frame containing the factor variables. Must be
factor s |
mc.matrix |
a list of matrices giving the probabilities for the
misclassification. Names of the list must correspond to the variable
names in data.org . The colnames must be named according to the
factor levels |
k |
the exponent for the misclassification matrix |
A data.frame
containing the misclassified variables
Wolfgang Lederer, wolfgang.lederer@googlemail.com
mcsimex
, build.mc.matrix
,
diag.block
x1 <- factor(rbinom(100, 1, 0.5)) x2 <- factor(rbinom(100, 2, 0.5)) p1 <- matrix(c(1, 0, 0, 1), nrow = 2) p2 <- matrix(c(0.8, 0.1, 0.1, 0.1, 0.8, 0.1, 0.1, 0.1, 0.8), nrow = 3) colnames(p1) <- levels(x1) colnames(p2) <- levels(x2) x <- data.frame(x1 = x1, x2 = x2) mc.matrix <- list(x1 = p1, x2 = p2) x.mc <- misclass(data.org = x, mc.matrix = mc.matrix, k = 1) identical(x[, 1], x.mc[, 1]) # TRUE identical(x[, 2], x.mc[, 2]) # FALSE