alpha.continuous.cond {gnonadd}

variance effect conditioned on continuous variables

Description

We estimate the variance effect of a variant conditioned on one or more continous variable We apply a likelyhood ratio test with one degree of freedom

H0: All covariates have a variance effect, but not the variant H1: The variant has a variance effect, and the covariates as well

Usage

alpha.continuous.cond(qt, g, x, iter_num = 50, eps_param = 1e-10)

Arguments

qt

A numeric vector.

g

An integer vector.

x

A numeric matrix, each column represents a covariate.

iter_num

An integer. Represents the number of iterations performed in the Gauss-Newton algorithm

eps_param

A number. The Gauss-Newton algorithm terminates if the incriment change of all variance estimates is smaller than this number.

Value

A list with the values: * alpha, the estimated variance effect, conditioned on the covariates * pval, the p-value corresponding to alpha

Examples

n_val <- 50000
x <- matrix(0,nrow = n_val, ncol = 4)
for(i in 1:4) {
x[, i] <- rnorm(n_val)
}
g_vec <- rbinom(n_val,2,0.3)
var_vec <- exp(0.2 * x[, 1] - 0.3 * x[, 4] + 0.3 * g_vec)
qt_vec <- rnorm(n_val, 0, sqrt(var_vec))
res <- alpha.continuous.cond(qt_vec, g_vec, x)