[R] (no subject)

[Ben Bolker]

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