xSJpearsonPMF {SimJoint} | R Documentation |
Simulate joint with marginal PMFs, Pearson correlations and uncorrelated support matrix.
Description
Sample from marginal probability mass functions via Latin hypercube sampling and then simulate the joint distribution with Pearson correlations. Users specify the uncorrelated random source instead of using permuted marginal samples to left-multiply the correlation matrix decomposition.
Usage
xSJpearsonPMF(
PMFs,
sampleSize,
cor,
noise,
stochasticStepDomain = as.numeric(c(0, 1)),
errorType = "meanSquare",
seed = 123L,
maxCore = 7L,
convergenceTail = 8L,
iterLimit = 100000L,
verbose = TRUE
)
Arguments
PMFs |
A list of data frames. Each data frame has 2 columns, a value vector and a probability vector. Probabilities should sum up to 1. Let the size of |
sampleSize |
An integer. The sample size |
cor |
A |
noise |
An |
stochasticStepDomain |
A numeric vector of size 2. Range of the stochastic step ratio for correcting the correlation matrix in each iteration. Default [0, 1]. See the package vignette for more details. |
errorType |
Cost function for convergence test.
|
seed |
An integer or an integer vector of size 4. A single integer seeds a |
maxCore |
An integer. Maximal threads to invoke. Default 7. Better be no greater than the total number of virtual cores on machine. |
convergenceTail |
An integer. If the last |
iterLimit |
An integer. The maximal number of iterations. Default 100000. |
verbose |
A boolean value. |
Details
Algorithms are detailed in the package vignette.
Value
A list of size 2.
X |
A numeric matrix of size |
cor |
Pearson correlation matrix of |
Examples
# =============================================================================
# Use the same example from <https://cran.r-project.org/web/packages/
# SimMultiCorrData/vignettes/workflow.html>.
# =============================================================================
set.seed(123)
N = 10000L # Sample size.
K = 10L # 10 marginals.
# 3 PDFs, 2 nonparametric PMFs, 5 parametric PMFs:
PMFs = c(
apply(cbind(rnorm(N), rchisq(N, 4), rbeta(N, 4, 2)), 2, function(x)
data.frame(val = sort(x), P = 1.0 / N)),
list(data.frame(val = 1:3 + 0.0, P = c(0.3, 0.45, 0.25))),
list(data.frame(val = 1:4 + 0.0, P = c(0.2, 0.3, 0.4, 0.1))),
apply(cbind(rpois(N, 1), rpois(N, 5), rpois(N, 10),
rnbinom(N, 3, 0.2), rnbinom(N, 6, 0.8)), 2, function(x)
data.frame(val = as.numeric(sort(x)), P = 1.0 / N))
)
# Create the target correlation matrix `Rey`:
set.seed(11)
Rey <- diag(1, nrow = 10)
for (i in 1:nrow(Rey)) {
for (j in 1:ncol(Rey)) {
if (i > j) Rey[i, j] <- runif(1, 0.2, 0.7)
Rey[j, i] <- Rey[i, j]
}
}
system.time({result = SimJoint::xSJpearsonPMF(
PMFs = PMFs, sampleSize = N, noise = matrix(runif(N * K), ncol = K),
cor = Rey, errorType = "meanSquare", seed = 456, maxCore = 1,
convergenceTail = 8, verbose = TRUE)})
# Check relative errors.
summary(as.numeric(abs(result$cor / Rey - 1)))