simu.sample.real: Simulate a sample to do t test

In warnes/exp.ssize: Estimate Effectivenes of Microarry Sample Size Estimation via Simulation

Description

Simulate a sample to do t test

Usage

simu.sample.real(sd.null, sd.alt, cov.null, ngenes.null.dep, cov.alt, ngenes.alt.dep, n, var.ratio, delta)

Arguments

sd.null	sd vector for observations that are independent and come from Ho
sd.alt	sd vector for observations that are independent and come from Ha
cov.null	list of covariance matrices for observations that are dependent and come from Ho
ngenes.null.dep	number of observations in dependent groups for observations coming from Ho
cov.alt	list of covariance matrices for observations that are dependent and come from Ha
ngenes.alt.dep	number of observations in dependent groups for observations coming from Ha
n	number of replicates to simulate for each group of control/treatment
var.ratio	ratio of variance between treatment/control
delta	effect size for sample size calculation the absolute distance from null value for expression values of genes coming from Ha group

Details

simulate observations from both control group and treatment group

Value

returns a matrix of data with each row for a observation and columns are replicates, observations from Ho group occupy first rows followed by observations from Ha first n columns are from control group and second n columns are from treatment group

Author(s)

Peng Liu [e-mail]

Examples

library(MASS)
sd.null <- rep(0.1,4); sd.alt <- rep(0.2,5);
cov.null <- cov.alt <- list(2)
cov.null[[1]] <- cov.alt[[1]] <- cbind(c(0.1, 0.05), c(0.05,0.2));
cov.null[[2]] <- cbind(c(0.15, 0.05), c(0.05,0.2));
cov.alt[[2]] <- cbind(c(0.039, -0.037, 0.018), c(-0.037, 0.045, -0.021),
	 c(0.018, -0.021, 0.013));

test <- simu.sample.real(sd.null, sd.alt, cov.null, ngenes.null.dep = 4,
	 cov.alt, ngenes.alt.dep=5, n = 15,	var.ratio = 1, delta = 5)

# dim(test) should be [18,30]

warnes/exp.ssize documentation built on May 4, 2019, 12:59 a.m.