d_par: d_par

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

Description

Distance metric between sets of parameters. Computed as a Euclidean distance in a transformed parameter space. pi0-pi2 are transformed to their inverse logits, as a difference 0.001-0.1 means more than a difference 0.4-0.5. Tau, sigma1, and sigma2 are transformed to sqrt(|x-1|), as differences are most important for values near 1. Rho is transformed to x^2 /2, since differences are only important for large rho.

Usage

d_par(pars1, ...)

Arguments

pars1	an n x 6 matrix of parameters
...	additional parameters passed to dist

Value

an n x m matrix A such that A[i,j] is the distance between pars1[i,] and pars2[j,]

Author(s)

James Liley

Examples

px=pars_start(N=5)
dx=d_par(px)
image(as.matrix(dx))

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