Skip to contents

Summary graph

summary_graph.Rd

Construct summary graph from p-values and significance level. Recursively constructs all ancestral connections by adding ancestors of ancestors.

Usage

summary_graph(lin.anc, alpha = 0.05, corr = TRUE)

Arguments

lin.anc

output from AncReg()

alpha

significance level

corr

should multiplicity correction be applied?

Value

A boolean matrix indicating whether one variable affects another

See also

AncReg, summary_p.val

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)

# generate summary graph
summary_graph(fit, alpha = 0.1)
#>       A     B     C     D     E
#> A FALSE FALSE FALSE FALSE FALSE
#> B FALSE FALSE FALSE FALSE FALSE
#> C FALSE FALSE FALSE FALSE FALSE
#> D FALSE FALSE FALSE FALSE FALSE
#> E FALSE FALSE FALSE FALSE FALSE