RclusTool (version 0.91.61)

computePcaNbDims: Number of dimensions for PCA

Description

Compute the number of dimensions to keep after a Principal Components Analysis, according to a threshold on the cumulative variance.

Usage

computePcaNbDims(sdev, pca.variance.cum.min = 0.9)

Value

pca.nb.dims number of dimensions kept.

Arguments

sdev

standard deviation of the principal components (returned from prcomp).

pca.variance.cum.min

minimal cumulative variance to retain.

Details

computePcaNbDims computes the number of dimensions to keep after a Principal Components Analysis, according to a threshold on the cumulative variance

See Also

computePcaSample

Examples

dat <- rbind(matrix(rnorm(100, mean = 0, sd = 0.3), ncol = 2), 
             matrix(rnorm(100, mean = 2, sd = 0.3), ncol = 2), 
             matrix(rnorm(100, mean = 4, sd = 0.3), ncol = 2))
tf <- tempfile()
write.table(dat, tf, sep=",", dec=".")

x <- importSample(file.features=tf)
res.pca <- computePcaSample(x)
computePcaNbDims(res.pca$pca$sdev)