Generate one-sided p-values associated to a given signal with equi-correlated noise — gen.p.values • sanssouci

Generate one-sided p-values associated to a given signal with equi-correlated noise

Usage

gen.p.values(
  m,
  mu = rep(0, m),
  rho = 0,
  alternative = c("two.sided", "less", "greater")
)

Arguments

m: An integer value, the number of hypotheses
mu: A vector of \(m\) signal values
rho: A numeric value in \([0,1]\), the level of equi-correlation between test statistics
alternative: A character string specifying the alternative hypothesis. Must be one of "two.sided" (default), "greater" or "less".

Value

A vector of \(m\) one-sided \(p\)-values

Examples

m <- 100
s <- 10
K1 <- floor(m/(s * 4))
d <- 1
barmu <- 4
dd <- dyadic.from.window.size(m, s, method = 2)
leaf_list <- dd$leaf_list
mu <- gen.mu.leaves(m = m, K1 = K1, d = d, grouped = FALSE, 
                    setting = "const", barmu = barmu, leaf_list =leaf_list)
pvals <- gen.p.values(m = m, mu = mu, rho = 0)
plot(-log(pvals), t = 'b')