how to apply a function on several simulated graphs in R

Question

I need a vector (or list) of several results of a function on several simulated graphs (eg assortativity). This is to emulate a Monte Carlo procedure, to perform a segregation analysis with stochastic block models instead of Erdos-Renyi models. This allows to differentiate the probability of links between two types of actors due to the design of the data collection, with a method similar to conditional uniform graphs. I have tried unsuccessfully with the following loop:

library(igraph)
k <- cbind(c(.2,.2),c(.2,0))  # one block without data of its links 
attrib <- c(1,1,2,2,2,2,2,2,2,2,3,3,3,3,3,5,5,5,5,5,5,5,5,5,1,1,2,2,2,2,2,2,2,3,3,4,4,4,4,4,4,5,5,5,5,5,5,5,5)  # observed data attributes for each block

for(i in 1:1000) {
  sbm[i] <- sample_sbm(49, pref.matrix=k,block.sizes=c(24,25))
  sbm[i] <- set_vertex_attr(sbm[i], "att",value=attrib)
  sbm_ass[i] <- assortativity_nominal(sbm[i],V(sbm[i])$att,directed=F)
 }

I've also tried with [i] only on the last line: "sbm_ass[i]...". I suppose that a function is needed, but I have not been able to realize it.

Answer 1

I think replicate is the thing you may need

sbm_ass <- replicate(
  1000,
  sample_sbm(49, pref.matrix = k, block.sizes = c(24, 25)) %>%
    set_vertex_attr(name = "att", value = attrib) %>%
    assortativity_nominal(V(.)$att, directed = F)
)