tindep: Standardized Scores for Two Independent Samples

Description

Calculate the independent two-sample Welch t-statistics for k populations simultaneously. This function is used in PCS.bootstrap.np. It may also be used to summarize data from two sample experiments for use in PCS.exact and PCS.bootstrap.par.

Usage

tindep(X, Y, flag = 0)

Arguments

k by n1 matrix of data. k is the number of populations and n1 the sample size of the first treatment.

k by n2 matrix of data. k is the number of populations and n2 the sample size of the second treatment.

flag

Determines whether the output is t-statistics (flag=0) or t-statistics with p-values (flag!=0)

Value

If flag = 0, the result is a k by 1 vector of t-statistics. If flag != 0, the result is a k by 7 vector of t-statistics and pvalues (see details).

Details

X & Y are the data matrices for input. When called inside of the function PCS.bootstrap.np, or when used to obtain t-statistics for use with PCS.exact and PCS.bootstrap.par, setting 'flag'=0 will return the needed vector of t-statistics. If flag!=0, then the function 'mt.rawp2adjp' from the multtest library is called, producing multiple comparison adjustd pvalues ("Bonferroni", "Holm", "Hochberg", "SidakSS", "SidakSD", "BH", "BY"). The t-statistics and p-values are stored in a matrix.

Examples

Run this code

# NOT RUN {
### Small example
k=20 #number of populations
n=10 #sample size
SD=.1 #standard deviation
theta = seq(0,6,length.out=k)
X1 = rnorm(k*n,0,SD)   #Sample 1
X1 = matrix(X1,nrow=k,ncol=n,byrow=FALSE)
X2 = rnorm(k*n,theta,SD) #Sample 2, shifted
X2 = matrix(X2,nrow=k,ncol=n,byrow=FALSE)
tindep(X1,X2, flag=1)

### Microarray example
#require(multtest)
data(golub)
sub = 500 #Subset index for speed
golub.Tp = tindep(golub[1:sub,1:27], golub[1:sub,28:38], flag=0) #Obtain t-statistics + p-values

ans = tindep(golub[1:sub,1:27], golub[1:sub,28:38], flag=1)  #Obtain t-statistics
golub.T = sort(abs(ans[,1]))
# }