P-values for summary graph

Collect p-values for summary graph.

Usage

summary_p.val(lin.anc)

Arguments

lin.anc

output from AncReg()

Value

A numeric matrix of p-values for the summary graph

See also

AncReg

Examples

# random DAGS for simulation
set.seed(1234)

p <- 5 #number of nodes
DAG <- pcalg::randomDAG(p, prob = 0.5)

B <- matrix(0, p, p) # represent DAG as matrix
for (i in 2:p){
  for(j in 1:(i-1)){
    # store edge weights
    B[i,j] <- max(0, DAG@edgeData@data[[paste(j,"|",i, sep="")]]$weight)
  }
}
colnames(B) <- rownames(B) <- LETTERS[1:p]

# solution in terms of noise
Bprime <- MASS::ginv(diag(p) - B)

n <- 500
N <- matrix(rexp(n * p), ncol = p)
X <- t(Bprime %*% t(N))
colnames(X) <- LETTERS[1:p]

# fit ancestor regression
fit <- AncReg(X)

# collect summary p-values
summary_p.val(fit)
#>           A           B         C         D         E
#> A 1.0000000 0.355083093 0.0493693 0.5107251 0.9777551
#> B 0.6612617 1.000000000 0.5350391 0.1672114 0.5645382
#> C 0.3190314 0.224148437 1.0000000 0.9467940 0.9968308
#> D 0.8740048 0.006213275 0.3530013 1.0000000 0.2938822
#> E 0.4328651 0.798593250 0.4182837 0.1692897 1.0000000