GenerateFoldMatrix {GeneSelector} | R Documentation |
Generates an object of class FoldMatrix to be used for RepeatRanking.
GenerateFoldMatrix(x, y, k = 1, replicates = ifelse(k==1, length(y), 10), type = c("unpaired", "paired", "onesample"), minclassize = 2, balanced = FALSE, control)
x |
Only needed if |
y |
|
k |
Number of observations that are removed or whose labels are exchanged. Label exchange means that the observed label is replaced by the label of the other class (s. RepeatRanking). |
replicates |
Number of replications if |
type |
One of |
minclassize |
If |
balanced |
If |
control |
Further control arguments concerning the generation process of the fold matrix, s. samplingcontrol. |
An object of class FoldMatrix.
If the generation process (partially) fails, try to
reduce the constraints or change the argument control
.
No jackknif-ed dataset will occur more than once, i.e. each replication is unique.
Martin Slawski
Anne-Laure Boulesteix
Davison, A.C., Hinkley, D.V. (1997)
Bootstrap Methods and their Application.
Cambridge University Press
GenerateBootMatrix, RepeatRanking
## Load toy gene expression data data(toydata) ### class labels yy <- toydata[1,] ### Generate Leave-One-Out / Exchange-One-Label matrix loo <- GenerateFoldMatrix(y = yy, k=1) ### A more complex example l3o <- GenerateFoldMatrix(y = yy, k=3, replicates=30, minclassize=5)