orQA (version 0.2.1)

guo: Mixed directional FDR controlled test decisions

Description

Compute mixed directional FDR controlled test decisions for raw p-values of multidimensional ordered tests

Usage

guo(p,a)

Arguments

p

a numeric matrix, of raw p-values from multidimensional ordered tests. Each line corresponds to a group of ordered hypotheses e.g. t-tests between mean mesurements from consecutive levels of some ordered grouping variable.

a

desired alpha level for the test

Value

A matrix with test decisions TRUE corresponds to significant rejection of the null and FALSE to acceptance of the null.

Details

guo implements the method proposed in Guo et al. (2010) that provides test decisions controlling the mixed directional FDR for multidimensional ordered tests.

References

Klinglmueller, F., Tuechler, T., Posch, M. (2010) "Cross Platform Comparison Of Microarray Data Using Order Restricted Inference" Under Review

Guo W., Sarkar SK., Peddada SD. (2010) "Controlling False Discoveries in Multidimensional Directional Decisions, with Applications to Gene Expression Data on Ordered Categories" Biometrics

Examples

data <- matrix(rnorm(7200),nc=72)
groups <- rep(1:4,each=18)
ind <- rep(rep(1:3,each=6),4)
out <- pttest(data,groups,B=1000,rep=ind)
guo(out$unadj,.05)
data2 <- data+matrix(rep(groups,nrow(data)),nr=nrow(data),byrow=TRUE)
out2 <- pttest(data2,groups,B=1000,rep=ind)
guo(out2$unadj,.05)