Plot region objects {cherry}

Visualizing of the region hypotheses that could be rejected.

Description

Visualizes region objects as created through a call to regionmethod.

Usage

  regionplot (region, alpha, color="red")
  
  regionplot2 (region, alpha, color_rej="red", color_unrej="grey") 

Arguments

region

An object of class region, typically created through a call to regionmethod.

alpha

For region objects with adjusted p-values, specifies the value of alpha for which rejections should be plotted (optional).

color

Color that is used to indicate rejected region hypotheses.

color_rej

Color that is used to indicate rejected region hypotheses.

color_unrej

Color that is used to indicate unrejected region hypotheses.

Details

Both plot functions create a graph that visualizes all possible region hypotheses. Each region hypothesis is a node in the graph, and from each region hypothesis two edged connect the hypothesis with its child hypotheses. The regionplot2 function visualized the graph with its nodes and edges. This function is especially useful for region objects with a limited number of elementary hypotheses. The regionplot function does not display the nodes and edges separately, but draws a polygon that follows the original graph structure.

Author(s)

Rosa Meijer: [e-mail]

Examples

#generate data, where the response Y is associated with certain groups of covariates
#namely cov 3-6, 9-12, 15-18
set.seed(1)
n=100
p=20
X <- matrix(rnorm(n*p),n,p)
beta <- c(rep(0,2),rep(1,4),rep(0,2),rep(1,4),rep(0,2),rep(1,4),rep(0,2))
Y <- X %*% beta + rnorm(n)

# Define the local test to be used in the closed testing procedure
mytest <- function(left,right)
{
  X <- X[,(left:right),drop=FALSE]
  lm.out <- lm(Y ~ X)
  x <- summary(lm.out)
  return(pf(x$fstatistic[1],x$fstatistic[2],x$fstatistic[3],lower.tail=FALSE))  
}

# perform the region procedure
reg <- regionmethod(rep(1,p), mytest, isadjusted=TRUE)
summary(reg)

#what are the smallest regions that are found to be significant? 
implications(reg)

#how many covariates within the full region of length 20 are at least associated with the response?
regionpick(reg, list(c(1,p)), alpha=0.05)

#visualize the results by either plotting a polygon corresponding to the underlying graph
regionplot(reg)

#or by plotting the graph itself
regionplot2(reg)