Creating Monte Carlo data for Dummy Variables in R - r

Question

I'm setting up an Monte Carlo simulation, and I have been trying to create a set of dummy variables for 180 countries and 12 time periods. Given the large amount of data points, is there a shorter way to create dummy variables for time and country fixed effects without pulling it out of an excel file?
For Example
F.T(1) 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0. 1, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0......(Extends until 180 countries)
F.T(2) 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0. 0, 1, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0......(Extrends until 180 countries)
Any help would be greatly appreciated.

Answer 1

Using replicate with your random number generator of choice should do the trick
Here is an example using a simple binomial distribution with prob = 1/2
replicate(12, rbinom(180, 1, .5), simplify=FALSE)

Answer 2

I think it might be easier/faster to create all data with rbinom first and then convert it into a matrix instead of calling rbinom 12 times. That is:
set.seed(45)
t <- rbinom(180*12, 1, 0.5)
dim(t) <- c(180, 12)
Just to see if there 's a difference, here's a benchmark
# I use simplify = TRUE here.
FUN1 <- function(n, a) {
set.seed(45)
replicate(n, rbinom(a, 1, .5), simplify = TRUE)
}
FUN2 <- function(n, a) {
set.seed(45)
t <- rbinom(n*a, 1, 0.5)
dim(t) <- c(a, n)
t
}
require(rbenchmark)
benchmark(t1 <- FUN1(1000, 12000), t2 <- FUN2(1000, 12000),
order="elapsed", replications=5)
# test replications elapsed relative user.self sys.self
# 2 t2 <- FUN2(1000, 12000) 5 3.991 1.000 3.859 0.111
# 1 t1 <- FUN1(1000, 12000) 5 5.337 1.337 4.785 0.472
identical(t1, t2)
# [1] TRUE
To answer your question in comment:
w <- rep(diag(12)[1:9, ], N)
dim(w) <- c(9, 12*N)
w <- t(w)
colnames(w) <- paste0("t", 1:9)
Or even better:
w2 <- do.call(rbind, replicate(N, diag(12)[, 1:9], simplify = FALSE))
colnames(w2) <- paste0("t", 1:9)