The rand_bytes
function binds to RAND_bytes in OpenSSL to generate cryptographically strong pseudo-random bytes. See the OpenSSL documentation for what this means.
rnd <- rand_bytes(10)
print(rnd)
[1] bc 17 8a 75 3f 55 a0 05 97 67
Bytes are 8 bit and hence can have 2^8 = 256
possible values.
as.numeric(rnd)
[1] 188 23 138 117 63 85 160 5 151 103
Each random byte can be decomposed into 8 random bits (booleans)
x <- rand_bytes(1)
as.logical(rawToBits(x))
[1] TRUE TRUE FALSE TRUE FALSE TRUE FALSE FALSE
rand_num
is a simple (2 lines) wrapper to rand_bytes
to generate random numbers (doubles) between 0 and 1.
rand_num(10)
[1] 0.1535481 0.5023007 0.8202547 0.6546440 0.4158769 0.4949917 0.4841464
[8] 0.9379224 0.3792755 0.6939537
To map random draws from [0,1] into a probability density, we can use a Cumulative Distribution Function. For example we can combine qnorm
and rand_num
to simulate rnorm
:
# Secure rnorm
x <- qnorm(rand_num(1000), mean = 100, sd = 15)
hist(x)
Same for discrete distributions:
# Secure rbinom
y <- qbinom(rand_num(1000), size = 20, prob = 0.1)
hist(y, breaks = -.5:(max(y)+1))