plhood1: plhood1

In jamesliley/subtest: Test whether two putative subgroups of a disease phenotype show evidence of differential genetic basis.

Description

Computes the likelihood of observed Z_a.

Usage

plhood1(za, pars, weights = rep(1, length(za)), C = 1, sumlog = TRUE)

Arguments

za	one dimensional vector of Z_a scores
pars	vector of parameters (pi0,pi1,tau,sigma1,sigma2,rho)
weights	LDAK weights corresponding to Z_a
C	scaling factor for adjustment
sumlog	set to TRUE to return the (weighted) sum of log pseudo-likelihoods for each datapoint; FALSE to return a vector of pseudo-likelihoods for each datapoint.

Details

Used for establishing how much of a likelihood ratio is because Z_a is better-fitted by a three-Gaussian model than a two-Gaussian model. Strictly, calculates the expected value of PL(Z_d,Z_a) when Z_d has uniform unit variance, and the parameter values tau, rho are set to 1, 0 respectively.

Value

value of pseudo- log likelihood of all observations (if sumlog==TRUE) or vector of n pseudo-likelihoods (if sumlog==FALSE)

Author(s)

James Liley

jamesliley/subtest documentation built on May 18, 2019, 11:21 a.m.