Simulate correlated background effects.

correlatedBgEffects computes a background effect that simulates structured correlation between the phenotypes.

correlatedBgEffects(N, P, pcorr = NULL, corr_mat = NULL,
  sampleID = "ID_", phenoID = "Trait_", id_samples = NULL,
  id_phenos = NULL, verbose = FALSE)

Arguments

N	Number [integer] of samples to simulate.
P	Number [integer] of phenotypes to simulate.
pcorr	Initial strength of correlation [double] between neighbouring traits. Decreases by pcorr^(distance); distance from 0 to P-1. See details.
corr_mat	[P x P] correlation matrix [double] as covariance component for the multivariate normal distribution. If not provided, pcorr is used to construct the correlation matrix.
sampleID	Prefix [string] for naming samples.
phenoID	Prefix [string] for naming traits.
id_samples	Vector of [NrSamples] sample IDs [string]; if not provided constructed by paste(sampleID, 1:N, sep="").
id_phenos	Vector of [NrTraits] phenotype IDs [string]; if not provided constructed by paste(phenoID, 1:P, sep="").
verbose	[boolean] If TRUE, progress info is printed to standard out.

Value

Named list with [N x P] matrix of correlated background effects ( correlatedBg) and the correlation matrix (cov_correlated). If corr_mat provided corr_mat == cov_correlated.

Details

correlatedBgEffects can be used to simulate phenotypes with a defined level of correlation between traits. If the corr_mat is not provided, a simple correlation structure based on the distance of the traits will be constructed. Traits of distance d=1 (adjacent columns) will have correlation cor=\(pcorr^1\), traits with d=2 have cor=\(pcorr^2\) up to traits with d=(P-1) cor=\(pcorr^{(P-1)}\) and 0 < pcorr < 1. The correlated background effect correlated is simulated based on this correlation structure C: \(correlated ~ N_{NP}(0,C)\).

See also

rmvnorm which is used to simulate the multivariate normal distribution

Examples

correlatedBg <- correlatedBgEffects(N=100, P=20, pcorr=0.4)