How can I create a random dummy variable of 2000 observations in R and Stata?

Question

I want to create a random dummy (1 and 0) variable in R or Stata, but how can I make that, for example, 70% of observations be 1 and the rest 0. Thanks

Answer 1

Here's an approach with sample from base R:

sample(c(1,0), size = 2000, prob = c(0.7,0.3), replace = TRUE)
# [1] 1 1 1 1 0 1 1 1 1 1 1 0 1 1 1 1 1 0 1 0 1 1 1 1 0 0 1 1 1 1 0 1 1 1 1 0 0 0 0 0 1 0 0 1 1 0 1 1 1 1 1 0 0 1 1 1 1
#[58] 1 1 1 1 0 1 1 0 1 0 1 0 1 0 1 1 1 0 1 1 0 1 0 1 1 1 1 0 1 1 1 1 0 1 1 1 1 1 0 1 1 0 0 1 0 1 1 0 1 1 1 1 1 1 1 1 1

As @Ben Bolker points out in the comments, it would be unusual for exactly 1400 to be 1 .

This approach will result in exactly 1400 1 s:

sample(rep(c(1,0),c(1400,600)), 2000)

Answer 2

If you wanted exactly 70% of 1s (or any other percentage), but a random ordering of the elements, you can use this function.

random_binary <- function(n, p){
  # p is the proportion of 1s
  x <- c(rep(1, times=n * p), rep(0, times=n * (1 - p)))
  x[sample(length(x))] # or sample(x)
}

random_binary(10, 0.7)
#[1] 1 0 1 1 0 0 1 1 1 1

The times argument of rep can be non-integer, as mentioned in the documentation.

? rep

times

A double vector is accepted, other inputs being coerced to an integer or double vector.

But note that you may not get exactly the percentage desired (but as close as possible).

An alternative is to use rbinom , since we're effectively sampling from the binomial distribution.

rbinom(10, size=1, p=0.7)
# [1] 0 0 0 0 1 1 1 0 1 0

This is similar to sample with the prob argument and, as shown above, does not guarantee to return exactly 70% of 1s.

Answer 3

In Stata for exactly 70% 1s and 30% 0s

set obs 2000
set seed 1606
gen wanted = cond(_n <= 70, 1, 0) 
gen random = runiform()
sort random 

For approximately 70% 1s and 30% 0s

gen better = runiform() < 0.7