Philip Twumasi-Ankrah <nana_kwadwo_derkyi <at> yahoo.com> writes:
> > Teds reply is a bit comforting and as indicated in my post, > I am resorting to > using "sample" but as an academic > issue, does randomness preclude precision? > > Randomness should be in the sequence of zeros and ones and > how they are simulated at each iteration of the > process but not in the eventual nature of the distribution. > > I mean if I simulated a Normal (0, 1) and got a Normal(1.5, 2) > these would be > very different distributions. It > is the same with simulating a Binomial(1, p=0.15) and getting > Binomial(1, 0.154) > It's impossible to have a sampler that is (1) Binomial(1,p) on each draw and (2) independent for each draw and (3) has exactly Np successes in N draws. For example, suppose you had drawn 99 Binomial(1,p=0.15) samples and had got(ten) 14 successes so far ... your last draw would be constrained to be 1 if you wanted property #3 to hold. So I guess the answer to your question is "yes" (except in the limit of infinitely large samples). Ben Bolker ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.