fit.2g: fit.2g

In jamesliley/cfdr: Implements cFDR testing with overall FDR control

Description

Fit a specific two Guassian mixture distribution to a set of P or Z values.

Usage

fit.2g(P, pars = c(0.5, 1.5), weights = rep(1, min(length(Z),
  dim(Z)[1])), sigma_range = c(1, 100), rho = 0, ...)

Arguments

P	numeric vector of observed data, either p-values or z-scores. If rho=0, should be one-dimensional vector; if rho is set, should be bivariate observations (P,Q)
pars	initial values for parameters
weights	optional weights for parameters
sigma_range	range of possible values for sigma (closed interval). Default [1,100]

Details

Assumes Z ~ N(0,1) with probability pi0, Z ~ N(0,sigma^2) with probability 1-pi0

Returns MLE for pi0 and sigma. Uses R's optim function. Can weight observations.

Value

a list containing parameters pars, likelihoods under h1 (Z distributed as above), likelihood under h0 (Z~N(0,1)) and likelihood ratio lr.

Author(s)

James Liley

Examples

sigma=2; pi0 <- 0.8

n=10000; n0=round(pi0*n); n1=n-n0
Z = c(rnorm(n0,0,1),rnorm(n1,0,sqrt(1+ (sigma^2))))
fit=fit.2g(Z)
fit$pars

jamesliley/cfdr documentation built on July 31, 2020, 9:42 a.m.