-
Notifications
You must be signed in to change notification settings - Fork 1
/
01_ggm_network.R
168 lines (140 loc) · 7.02 KB
/
01_ggm_network.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
library(future)
plan(multisession)
load("data/ISSP_NI_M23_comp_miss_ranger.RData")
colnames(d)
library(tidyverse)
#unique(d$cntr)
#sample_df_long <- d %>% group_by(cntr, time) %>% summarise(COUNT = n())
#sample_df_wide <- sample_df_long %>% spread(time, COUNT)
#rio::export(sample_df_wide, file = "sample_df_wide.xlsx")
library(bootnet)
ni_nw_ggm <- estimateNetwork(d[,12:41], default = "EBICglasso", tuning = 0.5, missing = "pairwise", threshold = T, lambda.min.ratio = 0.01)
library(qgraph)
ni_nw_centralityPlot <- centralityPlot(ni_nw_ggm, scale = "z-scores", include = c("Strength","ExpectedInfluence"),
orderBy = "ExpectedInfluence")
plot(ni_nw_ggm, layout = "spring", maximum=1, edge.labels=T, edge.label.cex=1, vTrans = 200,
title = "Force-directed plotting with Fruchterman-Reingold")
ni_nw_wmat <- getWmat(ni_nw_ggm)
library(igraph)
ni_nw_gam <- graph_from_adjacency_matrix(abs(ni_nw_wmat), "undirected", weighted = T, add.colnames = F)
#################################################################
# community detection with louvain
ni_nw_comm_lo <- cluster_louvain(ni_nw_gam)
ni_nw_comm_lo_int <- ni_nw_comm_lo$membership; ni_nw_comm_lo_int
# 3 1 1 1 1 1 1 1 1 3 3 3 4 2 2 2 2 2 2 2 2 2 2 4 4 4 4 4 4 4
ni_nw_comm_lo_int <- c("3", "1", "1", "1", "1", "1", "1", "1", "1",
"3", "3", "3", "4", "2", "2", "2", "2", "2",
"2", "2", "2", "2", "2", "4", "4", "4", "4",
"4", "4", "4")
ni_nw_comm_graph_lo <- plot(ni_nw_ggm, layout = "spring", palette = "colorblind",
groups = ni_nw_comm_lo_int, legend=T, maximum=1, edge.labels=T, edge.label.cex=1,
details = T, title = "Force-directed plotting with Fruchterman-Reingold")
ni_nw_comm_lo_chr <- c("comparative",
"CoN", "CoN", "CoN", "CoN", "CoN", "CoN", "CoN", "CoN",
"comparative", "comparative", "comparative", "out",
"NatPri", "NatPri", "NatPri", "NatPri", "NatPri", "NatPri", "NatPri", "NatPri", "NatPri", "NatPri",
"out", "out", "out", "out", "out", "out", "out")
#par(mfrow=c(1,2))
ni_nw_comm_graph_lo <- plot(ni_nw_ggm, layout = "spring",
groups = ni_nw_comm_lo_chr, legend = T, maximum = 1, edge.labels = T, edge.label.cex = 1,
details = T, title = "Force-directed plotting with Fruchterman-Reingold")
Names <- scan("data/ISSP_NI_M23.RData_items_leg.txt", what = "character", sep = "\n")
nw_graph <- plot(ni_nw_ggm, layout = ni_nw_comm_graph_lo$layout,
legend=T, maximum=1, edge.labels=F,
nodeNames = Names, legend.cex = 0.3, vsize = 5, details = F, title = "National Identity Attitude Network // Node Labels in Legend")
#dev.off()
ni_nw_eigen <- ni_nw_wmat
diag(ni_nw_eigen) <- NA
p <- 2
library(eigenmodel)
fitEM <- eigenmodel_mcmc(Y = ni_nw_eigen, R = p, S = 10000, burn = 200, seed = 123)
EVD <- eigen(fitEM$ULU_postmean)
evecs <- EVD$vec[, 1:p]
round(colMeans(evecs),3)
ni_nw_comm_graph_lo_for_int_eigen <- plot(ni_nw_ggm, layout = evecs,
groups = ni_nw_comm_lo_chr, legend = F, maximum = 1, edge.labels = T, edge.label.cex = 1,
vTrans = 200,
details = T,
title = "Eigenmodel configuration | same network plotted on a 2-D latent space so that the x- & y- axis are interpretable")
round(smallworldIndex(ni_nw_comm_graph_lo)$index,2) # 1.15
qgraph::smallworldness(ni_nw_gam, B = 1001, up = 0.995, lo = 0.005) # 1.0487516
nw_igo <- as.igraph(ni_nw_comm_graph_lo, attributes=T)
round(graph.density(nw_igo, loops=F),2) # 0.5
round(igraph::transitivity(nw_igo),2) # 0.53
round(average.path.length(nw_igo),2) # 1.5
round(centralization.degree(nw_igo)$centralization,2) # 0.16
round(centralization.closeness(nw_igo)$centralization,2) # 0.16
round(centralization.betweenness(nw_igo)$centralization,2) # 0.03
simulationnw <- lapply(1:1001, erdos.renyi.game,
n = length(V(nw_igo)),
p.or.m = length(E(nw_igo)), type="gnm")
simulationnw.c <- sapply(simulationnw, igraph::transitivity)
simulationnw.l <- sapply(simulationnw, average.path.length)
hist(simulationnw.c, breaks=20)
hist(simulationnw.l, breaks=20)
plot(density(simulationnw.c))
plot(density(simulationnw.l))
simulationnw.c.alpha <- quantile(simulationnw.c, probs=c(.025, .975))
simulationnw.c.alpha
simulationnw.l.alpha <- quantile(simulationnw.l, probs=c(.025, .975))
simulationnw.l.alpha
par(mfrow=c(1,2))
plot(density(simulationnw.c), main="Clustering", xlim=c(.35,.6))
abline(v=simulationnw.c.alpha, col="green")
abline(v=igraph::transitivity(nw_igo), col="red", lty=2)
plot(density(simulationnw.l), main="Mean geodesic", xlim = c(1.4,1.94))
abline(v=simulationnw.l.alpha, col="green")
abline(v=average.path.length(nw_igo), col="red", lty=2)
# judgind by the simulation of a 1001 random networks
# the clustering coeffient & average short lenght values of the network
# is significantly different from what we would expect from random networks.
dev.off()
###Testing Connectivity
#### SW Index von Dalege 2015: https://surfdrive.surf.nl/files/index.php/s/w2Zfq0ArRCNQIOt
#### More about ASPL: https://www.researchgate.net/publication/316617763_A_Network_Perspective_on_Political_Attitudes_Testing_the_Connectivity_Hypothesis
#### More on Shortest path length: https://cran.r-project.org/web/packages/qgraph/qgraph.pdf
SW_Index <- function (Graph, ci = c (.1, .05, .01, .001))
{
randomC <- vector (, 1001)
randomL <- vector (, 1001)
for (i in 1:1001)
{
Rgraph <- erdos.renyi.game (vcount (Graph), ecount (Graph), 'gnm')
randomC [i] <- igraph::transitivity (Rgraph, 'average')
randomL [i] <- average.path.length(Rgraph)
}
MrandomC <- mean (randomC)
MrandomL <- mean (randomL)
Clustering.Graph = igraph::transitivity (Graph, 'average')
ASPL.Graph = average.path.length (Graph)
Index <- (Clustering.Graph / MrandomC) / (ASPL.Graph / MrandomL)
sm_sample <- vector (, 1001)
for (i in 1:1001)
{
Rgraph <- erdos.renyi.game (vcount (Graph), ecount (Graph), 'gnm')
sm_sample [i] <- (igraph::transitivity (Rgraph, 'average') / MrandomC) /(average.path.length(Rgraph) / MrandomL)
}
CI <- as.vector (((quantile (sm_sample, 1 - (ci / 2)) - quantile (sm_sample, ci / 2)) / 2) + 1)
return (list (SW.Index = Index, Upper.CI = data.frame (CI = ci, Value.CI = CI),
Clustering.Graph = Clustering.Graph, Clustering.Random.Graph = MrandomC,
ASPL.Graph = ASPL.Graph, ASPL.Random.Graph = MrandomL))
}
SW_Index (ni_nw_gam)
# values are going to be different slightly every time; changes in the 3rd decimal
#https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2323569/
#$SW.Index
#[1] 1.096617
#$Upper.CI
#CI Value.CI
#1 0.100 1.040656
#2 0.050 1.050077
#3 0.010 1.067866
#4 0.001 1.076460
#$Clustering.Graph
#[1] 0.5475522
#$Clustering.Random.Graph
#[1] 0.4993438
#$ASPL.Graph
#[1] 1.501149
#$ASPL.Random.Graph
#[1] 1.50125