Displaying 2 results from an estimated 2 matches for "sample_int".
2018 Sep 21
1
Bias in R's random integers?
...onvenience, but I have tried to keep the C++ low.
Interesting results:
The results for "simple" sampling are the same.
> set.seed(42)
> sample.int(6, 10, replace = TRUE)
[1] 6 6 2 5 4 4 5 1 4 5
> sample.int(100, 10)
[1] 46 72 92 25 45 90 98 11 44 51
> set.seed(42)
> sample_int(6, 10, replace = TRUE)
[1] 6 6 2 5 4 4 5 1 4 5
> sample_int(100, 10)
[1] 46 72 92 25 45 90 98 11 44 51
But there is no bias with the alternative method:
> m <- ceiling((2/5)*2^32)
> set.seed(42)
> x <- sample.int(m, 1000000, replace = TRUE)
> table(x %% 2)
0...
2018 Sep 20
5
Bias in R's random integers?
On 9/20/18 1:43 AM, Carl Boettiger wrote:
> For a well-tested C algorithm, based on my reading of Lemire, the unbiased
> "algorithm 3" in https://arxiv.org/abs/1805.10941 is part already of the C
> standard library in OpenBSD and macOS (as arc4random_uniform), and in the
> GNU standard library. Lemire also provides C++ code in the appendix of his
> piece for both this and