PoolBal (version 0.1-0)

hrPool: Empirical UMP beta pooled p-value

Description

Uses simulation under the null to approximate the UMP pooled p-value for a restricted beta family.

Usage

hrPool(w = 1, M = 10, nsim = 1e+05)

Value

A closure which accepts a vector of values between 0 and 1 and returns a single numeric between 0 and 1

Arguments

w

numeric value between 0 and 1

M

integer, the number of tests to pool

nsim

integer, the number of simulated null cases generated

Author

Chris Salahub

Details

To test the null hypotheses that all p-values are uniform against a restricted beta family 0 < a <= 1 <= b, the most powerful pooled p-value linearly combines upper and lower tail probabilities of the chi-squared distribution with two degrees of freedom with weights w and (1 - w) where w = (1 - a)/(b - a).

This function computes the statistic given by this combination for a collection of p-values, and then simulates a specified number of null cases to give an empirical pooled p-value. It produces a closure so that the time-intensive simulation step doesn't need to be repeated.

Examples

p <- c(0.1, 0.5, 0.9)
hr2 <- hrPool(w = 0.2, M = 3)
hr2(p)
hr5 <- hrPool(w = 0.5, M = 3, nsim = 100)
hr5(p)