jsd: Jensen-Shannon Distance

In hfgolino/EGA: Exploratory Graph Analysis – a Framework for Estimating the Number of Dimensions in Multivariate Data using Network Psychometrics

Jensen-Shannon Distance

Description

Computes the Jensen-Shannon Distance between two networks

Usage

jsd(network1, network2, method = c("kld", "spectral"), signed = TRUE)

Arguments

network1	Matrix or data frame. Network to be compared
network2	Matrix or data frame. Second network to be compared
method	Character (length = 1). Method to compute Jensen-Shannon Distance. Defaults to "spectral". Available options: "kld" — Uses Kullback-Leibler Divergence "spectral" — Uses eigenvalues of combinatorial Laplacian matrix to compute Von Neumann entropy
signed	Boolean. (length = 1). Should networks be remain signed? Defaults to TRUE

Value

Returns Jensen-Shannon Distance

Author(s)

Hudson Golino <hfg9s at virginia.edu> & Alexander P. Christensen <alexander.christensen at Vanderbilt.Edu>

Examples

# Obtain wmt2 data
wmt <- wmt2[,7:24]

# Set seed (for reproducibility)
set.seed(1234)

# Split data
split1 <- sample(
  1:nrow(wmt), floor(nrow(wmt) / 2)
)
split2 <- setdiff(1:nrow(wmt), split1)

# Obtain split data
data1 <- wmt[split1,]
data2 <- wmt[split2,]

# Perform EBICglasso
glas1 <- EBICglasso.qgraph(data1)
glas2 <- EBICglasso.qgraph(data2)

# Spectral JSD
jsd(glas1, glas2)
# 0.1595893

# Spectral JSS (similarity)
1 - jsd(glas1, glas2)
# 0.8404107

# Jensen-Shannon Divergence
jsd(glas1, glas2, method = "kld")
# 0.1393621

hfgolino/EGA documentation built on Nov. 11, 2024, 9:28 p.m.