epiGWAS (version 1.0.2)

normalized_outcome: Implements the normalized modified outcome approach

Description

Normalized modified outcome is an improvement to modified_outcome. Its large-sample variance is lower than the original modified outcome approach. The only difference between the two methods lies in the normalization of the propensity scores. The inverses of the propensity scores \(1/\pi(A=1| X)\) and \(1/\pi(A=0| X)\) are respectively normalized by their sum \(\sum_{i} 1/\pi(A_i=1| X_i)\) and \(\sum_{i} 1/\pi(A_i=0| X_i)\).

Usage

normalized_outcome(A, X, Y, propensity, parallel = FALSE, ...)

Arguments

A

target variant

X

rest of the genotype

Y

phenotype

propensity

propensity scores

parallel

whether to perform support estimation in a parallelized fashion

...

additional arguments to be passed to stabilityGLM or stabilityBIG

Value

a vector containing the area under the stability selection path for each variable in X

Examples

# NOT RUN {
n <- 30
p <- 10
X <- matrix((runif(n * p) < 0.5) + (runif(n * p) < 0.5),
             ncol = p, nrow = n) # SNP matrix
A <- (runif(n) < 0.3)
propensity <- runif(n, min = 0.4, max = 0.8)
Y <- runif(n) < 0.4
normalized_scores <- normalized_outcome(A, X, Y, propensity,
                               lambda_min_ratio = 0.02 , n_subsample = 1)

# }