Re: [Bugs-r-devel] [Fwd: [R2WinBUGS] bugs]

[Andrew G. <ge...@st...>]

I have only one suggestion.  Jouni's idea is good, but n.sims should be 
the total number of eventual iterations kept.  This is useful for 
post-processing.
Thus, I suggest that Jouni's new parameter should be called 
n.sims.target or n.sims.approx or something like that, to indicate that 
it represents the number of sims that you are trying to get, 
approximately.  Then
n.thin = max(1, floor(n.chains * (n.iter - n.burnin)/n.sims.approx))
and n.sims is defined as already done in R2WinBUGS.
Andrew


>
>
>
> -------- Original Message --------
> Subject: [R2WinBUGS] bugs
> Date: Thu, 20 Sep 2007 12:02:41 +0200
> From: jou...@no...
> To: li...@st...
>
> Hi Uwe
>
> There's a slight bug in R2WinBUGS (2.0.4 and apparently also for 2.1.6).
> The n.thin parameter default value is defined as
>
>         n.thin = max(1, floor(n.chains * (n.iter - n.burnin)/1000)),
>
> which restricts the number of simulations to keep to 1000. We should have
> a parameter
>
>         n.sims = 1000,
>
> and redefine n.thin as
>
>         n.thin = max(1, floor(n.chains * (n.iter - n.burnin)/n.sims)),
>
> to allow changing the number of simulations to keep, easily.
>
>
> [.....]
>
> regards
>
> Jouni

-- 
Andrew Gelman
Professor, Department of Statistics
Professor, Department of Political Science
Director, Applied Statistics Center
Columbia University, New York
ge...@st...
www.stat.columbia.edu/~gelman

Usual schedule:

Mon, Tues, Fri in Statistics Dept ofc:
  Social Work Bldg (Amsterdam Ave at 122 St), Room 1016
  phone 212-851-2142, fax 212-851-2164
Wed, Thurs in Political Science Dept ofc:
  International Affairs Bldg (Amsterdam Ave at 118 St), Room 731
  phone 212-854-7075, fax 212-222-0598
Wed, Thurs from 2:30pm onward in Playroom:
  International Affairs Bldg, Room 707

I'm out of town 11-12 Oct, 26 Oct, and 15-23 Nov.