eigenRef {FactoInvestigate}

Reference eigen value

Description

Compute the eigen values of random datasets, with the hypothesis of independence.

Usage

eigenRef(res, dim = NULL, q = 0.95, time = "10000L", parallel = TRUE)

Arguments

res

an object of class PCA, CA or MCA

dim

a numerical vector giving the factorial dimensions for with to compute the eigenvalues calculation.

q

the quantile of computed values to use as reference value (ie. the confidence about the signification of dimensions)

time

a character indicating the loop condition. This string is made of a number and a letter coupled. The number X with letter L means to compute X datasets exactly. The number X with letter s means to compute as many datasets as possible during approximativley X seconds.

parallel

a boolean : if TRUE, the computation uses map reduce on the processor cores to increase the performance. Useful for huge datasets

Value

datasets

the number of random datasets simulated.

quantile

the quantile used for the reference definition.

inertia

the reference inertia for the dimensions declared.

Author(s)

Simon Thuleau and Francois Husson

See Also

dimRestrict, inertiaDistrib

Examples

## Not run: 
data(decathlon)
res.pca = PCA(decathlon, quanti.sup = c(11:12), quali.sup = c(13), graph = FALSE)
eigenRef(res.pca, q = 0.95, time = "10s")

data(children)
res.ca = CA(children, row.sup = 15:18, col.sup = 6:8, graph = FALSE)
eigenRef(res.ca, q = 0.99, time = "10000L")

data(tea)
res.mca = MCA(tea, quanti.sup = 19,quali.sup = 20:36, graph = FALSE)
eigenRef(res.mca, dim = 1:8, q = 0.90, time = "10s")

## End(Not run)