networktools (version 1.1.0)

edge.impact: Edge Impact

Description

Generates a matrix of edge impacts for each specified node. Each scalar in a given matrix represents the degree to which the level of a node impacts the strength of a specified edge in the network

Usage

edge.impact(input, gamma, nodes = c("all"), binary.data = FALSE,
  weighted = TRUE, split = c("median", "mean", "forceEqual", "cutEqual",
  "quartiles"))

Arguments

input

a matrix or data frame of observations (not a network/edgelist). See included example datasets depression and social.

gamma

the sparsity parameter used in generating networks. Defaults to 0.5 for interval data and 0.25 for binary data

nodes

indicates which nodes should be tested. Can be given as a character string of desired nodes (e.g., c("node1","node2")) or as a numeric vector of column numbers (e.g., c(1,2)).

binary.data

logical. Indicates whether the input data is binary

weighted

logical. Indicates whether resultant networks preserve edge weights or binarize edges.

split

method by which to split network given non-binary data. "median": median split (excluding the median), "mean": mean split, "forceEqual": creates equally sized groups by partitioning random median observations to the smaller group, "cutEqual": creates equally sized groups by deleting random values from the bigger group,"quartile": uses the top and bottom quartile as groups

Value

edge.impact() returns a list of class "edge.impact" which contains:

impact

a list of matrices. Each symmetric matrix contains the edge impacts for the given node

lo

a list of matrices. Each symmetric matrix contains the edge estimates for the given node's lower half

hi

a list of matrices. Each symmetric matrix contains the edge estimates for the given node's upper half

edgelist

a list of dataframes. Each dataframe contains an edgelist of edge impacts

Details

For an explanation of impact functions in general, see impact.

Edge impact is the change in an edge's value as a function of a given node. A separate edge impact value is calculated for each edge in the network.

It is highly useful to plot the edge impacts as if they were a network. Positive edges in the resultant graph can be interpreted as edges that were made more positive by the given node, and negative edges can be interpreted as edges that were made more negative by the given node.

The $hi and $lo output of edge.impact can also be used to quickly visualize the difference in network structure depending on node level (see examples).

Examples

# NOT RUN {
out <- edge.impact(depression[450:550,1:3], nodes="anhedonia")
# }
# NOT RUN {
out1 <- edge.impact(depression)
out2 <- edge.impact(depression, gamma=0.65,
        nodes=c("sleep_disturbance", "psychomotor_retardation"))
out3 <- edge.impact(social, nodes=c(1:6, 9), binary.data=TRUE)

summary(out1)
plot(out1, nodes="concentration_problems")

# Visualize edge impacts of psychomotor_retardation
# as a single network
plot(out1, nodes="psychomotor_retardation", type.edgeplot="single")

# Visualize the edge impacts of psychomotor_retardation
# as contrast between high and low
plot(out1, nodes="psychomotor_retardation", type.edgeplot="contrast")

# Extract the impact of psychomotor_retardation on a single edge
out1$impact[["psychomotor_retardation"]]["worthlessness", "fatigue"]

# Extract edge impacts of node Dan in edgelist format
out3$edgelist$Dan
# }
# NOT RUN {
# }