Parametric population samples with covariance or correlation matrices
Using a multivariate normal model, random populations are generated using the suplied covariance matrix. A statistic is calculated on the random population and compared to the statistic calculated on the original matrix.
MonteCarloStat( cov.matrix, sample.size, iterations, ComparisonFunc, StatFunc, parallel = FALSE )
cov.matrix |
Covariance matrix. |
sample.size |
Size of the random populations |
iterations |
Number of random populations |
ComparisonFunc |
Comparison functions for the calculated statistic |
StatFunc |
Function for calculating the statistic |
parallel |
if TRUE computations are done in parallel. Some foreach backend must be registered, like doParallel or doMC. |
Since this function uses multivariate normal model to generate populations, only covariance matrices should be used.
returns the mean repeatability, or mean value of comparisons from samples to original statistic.
Diogo Melo, Guilherme Garcia
cov.matrix <- RandomMatrix(5, 1, 1, 10)
MonteCarloStat(cov.matrix, sample.size = 30, iterations = 50,
ComparisonFunc = function(x, y) PCAsimilarity(x, y)[1],
StatFunc = cov)
#Calculating R2 confidence intervals
r2.dist <- MonteCarloR2(RandomMatrix(10, 1, 1, 10), 30)
quantile(r2.dist)
## Not run:
#Multiple threads can be used with some foreach backend library, like doMC or doParallel
##Windows:
#cl <- makeCluster(2)
#registerDoParallel(cl)
##Mac and Linux:
library(doParallel)
registerDoParallel(cores = 2)
MonteCarloStat(cov.matrix, sample.size = 30, iterations = 100,
ComparisonFunc = function(x, y) KrzCor(x, y)[1],
StatFunc = cov,
parallel = TRUE)
## End(Not run)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.