Score.Test.Overall: Implement the overall genetic effect test for H0:

In SPA3G: SPA3G: R package for the method of Li and Cui (2012)

Description

Score.Test.Overall returns results of overall genetic effect test, including score test statistic, estimated degree of freedom and scale parameter, and test p-value.

Usage

Score.Test.Overall(Y, K1, K2, K3)

Arguments

Y	numerical vector: quantitative phenotypes
K1	matrix: kernel matrix of the first gene.
K2	matrix: kernel matrix of the second gene.
K3	matrix: elementwise multiplication of K1 and K2

Value

Score	score statistic
df	estimated degree of freedom for the scaled chi-square
scale	estimated scale parameter for the scaled chi-square
p.value	test p-value

Examples

## The function is currently defined as
  function (Y, K1, K2, K3) 
  {
    b <- mean(Y)
    sig2 <- var(Y)
    U <- t(Y - b) %*% (K1 + K2 + K3) %*% (Y - b)/(2 * sig2)
    M <- (K1 + K2 + K3)
    e <- TRACE(PROJECT(M))/2
    c11 <- TT(PROJECT(K1), PROJECT(K1))
    c12 <- TT(PROJECT(K1), PROJECT(K2))
    c13 <- TT(PROJECT(K1), PROJECT(K3))
    c22 <- TT(PROJECT(K2), PROJECT(K2))
    c23 <- TT(PROJECT(K2), PROJECT(K3))
    c33 <- TT(PROJECT(K3), PROJECT(K3))
    COV <- matrix(c(c11, c12, c13, c12, c22, c23, c13, c23, c33), 
        3, 3)
    Its <- c(TRACE(PROJECT(K1)), TRACE(PROJECT(K2)), TRACE(PROJECT(K3)))
    correct_COV <- (COV - Its %*% t(Its)/(n - 1))/2
    Itt <- sum(correct_COV)
    k <- Itt/(2 * e)
    v <- 2 * e^2/Itt
    pvalue <- pchisq(U/k, df = v, lower.tail = FALSE)
    object <- list(Score = U, p.value = pvalue, df = v, scale = k)
    class(object) <- "Score Test: tau1=tau2=tau3=0"
    return(object)
  }

SPA3G documentation built on May 2, 2019, 11:12 a.m.