[Rd] Bias in R's random integers?
bbolker @ending from gm@il@com
Wed Sep 19 16:03:35 CEST 2018
On 2018-09-19 09:40 AM, David Hugh-Jones wrote:
> On Wed, 19 Sep 2018 at 13:43, Duncan Murdoch <murdoch.duncan using gmail.com>
>> I think the analyses are correct, but I doubt if a change to the default
>> is likely to be accepted as it would make it more difficult to reproduce
>> older results.
> I'm a bit alarmed by the logic here. Unbiased sampling seems basic for a
> statistical language. As a consumer of R I'd like to think that e.g. my
> bootstrapped p values are correct.
> Surely if the old results depend on the biased algorithm, then they are
> false results?
Balancing backward compatibility and correctness is a tough problem
here. If this goes into base R, what's the best way to do it? What was
the protocol for migrating away from the "buggy Kinderman-Ramage"
generator, back in the day? (Version 1.7 was sometime between 2001 and
I couldn't find the exact commit in the GitHub mirror: this is related ...
‘normal.kind’ can be ‘"Kinderman-Ramage"’, ‘"Buggy
Kinderman-Ramage"’ (not for ‘set.seed’), ‘"Ahrens-Dieter"’,
‘"Box-Muller"’, ‘"Inversion"’ (the default), or ‘"user-supplied"’.
(For inversion, see the reference in ‘qnorm’.) The
Kinderman-Ramage generator used in versions prior to 1.7.0 (now
called ‘"Buggy"’) had several approximation errors and should only
be used for reproduction of old results.
More information about the R-devel