Last active
March 9, 2019 16:42
-
-
Save benlistyg/2e90d1e62daeb334f8fae7debf699b06 to your computer and use it in GitHub Desktop.
How does the degree distribution of a FCD network differ from the "true" network of interest?
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# Fixed Choice Design Lunchtime Simulation | |
# 2019-02-06 | |
library(igraph) | |
library(dplyr) | |
library(ggplot2) | |
#Experimental Conditions | |
conds <- expand.grid( | |
c(150), | |
c(0.50), | |
c(NA,5), | |
stringsAsFactors = F) %>% | |
data.frame | |
#Single Function | |
fcd <- function(n_people,p_edge,top_n){ | |
graph_dat <- erdos.renyi.game( | |
n = n_people, | |
p.or.m = p_edge, | |
directed = F) %>% | |
as_edgelist %>% | |
as.data.frame %>% | |
arrange(V1) %>% | |
mutate( | |
n_connections = sequence(rle(.[,1])$lengths)) %>% | |
group_by(V1) | |
if(missing(top_n) | is.na(top_n)){ | |
graph_dat %>% | |
.[,c(1,2)] %>% | |
graph_from_data_frame(.,directed = F) %>% | |
degree_distribution %>% | |
matrix(nrow=1) %>% | |
as.data.frame | |
} else if(!missing(top_n)){ | |
graph_dat %>% | |
top_n(-top_n) %>% | |
.[,c(1,2)] %>% | |
graph_from_data_frame(.,directed = F) %>% | |
degree_distribution %>% | |
matrix(nrow=1) %>% | |
as.data.frame | |
} | |
} | |
#Experimental Function | |
fcd_sim <- function(n_sims = 100, x, y, top_n){ | |
replicate(n = n_sims, expr = fcd(x,y,top_n)) | |
} | |
#Storing Results | |
sim_results <- mapply(fcd_sim, x = conds$Var1, y = conds$Var2, top_n = conds$Var3) %>% | |
apply(X = ., MARGIN = 2, FUN = function(x){t(do.call(bind_rows,x))}) | |
#Plotting over all results | |
lapply(sim_results[1], | |
FUN = function(x){ | |
matplot(rowMeans(x,na.rm = T), | |
type='l', | |
col = rainbow( | |
n = ncol(x), | |
s = 0.5)[15], | |
lty='solid', | |
lwd=5, | |
ylab = "", | |
xlab = "", | |
xlim = c(0,max(conds$Var1)), | |
ylim = c(0,max(conds$Var2,na.rm = T)),main = 'Avg. Degree Distribution for 100 random graphs\nP(Edge) = 0.5, Fixed Choice @ 5\n|V| = 150')}) | |
par(new = TRUE) | |
lapply(sim_results[2], | |
FUN = function(x) matplot(rowMeans(x,na.rm = T), | |
type='l', | |
col = rainbow( | |
n = ncol(x), | |
s = 0.5)[1], | |
lty='solid', | |
lwd=5, | |
ylab = "", | |
xlab = "", | |
xlim = c(0,max(conds$Var1)), | |
ylim = c(0,max(conds$Var2,na.rm = T)))) | |
nn <- ncol(conds[,-1]) | |
legend("right", | |
legend = c("FCD - Top 5 Connections","Full Network"), | |
col=c( | |
rainbow( | |
n = 100, | |
s = 0.5)[1], | |
rainbow( | |
n = 100, | |
s = 0.5)[15] | |
), | |
cex=0.8, | |
fill=c( | |
rainbow( | |
n = 100, | |
s = 0.5)[1], | |
rainbow( | |
n = 100, | |
s = 0.5)[15] | |
) | |
) | |
nn <- ncol(conds[,-1]) | |
legend("right", | |
legend = c("FCD - Top 5 Connections","'True' Network"), | |
col=c( | |
rainbow( | |
n = 100, | |
s = 0.5)[1], | |
rainbow( | |
n = 100, | |
s = 0.5)[15] | |
), | |
cex=0.8, | |
fill=c( | |
rainbow( | |
n = 100, | |
s = 0.5)[1], | |
rainbow( | |
n = 100, | |
s = 0.5)[15] | |
) | |
) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment