| alpha.calc {gnonadd} | R Documentation |
Genetic variance effects
Description
This function estimates the variance effect of a genetic variant on a quantitative trait. The genotype is coded as 0 (non-carrier), 1 (single copy of effect allele) and 2 (two copies of effect allele). The variance effect (alpha) is modelled to be multiplicitive. We use a likelihood ratio test with 1 degree of freedom, * H0: y~N(mu_g,sigma^2) * H1: y~N(mu_g,sigma^2\*alpha^g) Under the alternative model, the variance of the trait changes multiplicatively with genotype: Var(y|g=j)=alpha^j*sigma^2, j in 0,1,2
Usage
alpha.calc(qt, g)
Arguments
qt |
A numeric vector. |
g |
An integer vector. |
Value
A list with the values:
* alpha, the estimated variance effect * sigma2_alt, The variance estimated for non-carriers * pval, the p-value of the likelihood ratio test * chi2, the chi squared statistics (with one degree of freedom) corrisponding to the p-value
Examples
n_val <- 50000L
geno_vec <- sample(c(0, 1, 2), size = n_val, replace = TRUE)
qt_vec <- rnorm(n_val) * (1.1^geno_vec)
res <- alpha.calc(qt_vec, geno_vec)