[R] Retrieving samples when ussing BRugs.
ligges at statistik.tu-dortmund.de
Sat Dec 17 19:28:08 CET 2011
On 14.12.2011 10:44, Thiago Guerrera wrote:
> Hi Sacha,
> I was referring to the BRugs package, where I usually proceed as follows:
> result = BRugsFit (...)
> param_samples = samplesSample("param") (1)
> But (1) takes a long time for big chains. So my question was specific to
> BRugs, if there an alternative way to get the samples.
Short answer: No.
Long answer: Well, samplesHistory(..., plot=FALSE) may be more
convenient for several parameters at once, but it uses samplesSample
internally, that means: no.
> openbugs() mentioned by you seems to be from R2WinBUGS that I have never
> used. I'll try that. Thanks for the suggestion.
> On Fri, Dec 9, 2011 at 4:34 PM, Sacha Epskamp<sacha.epskamp at gmail.com>wrote:
>> If you obtain an output object with openbugs(), e.g.:
>> samples<- openbugs( ... )
>> Then this is a list with all the parameters and samples. See
>> names(samples). You can obtain the samples in a list with:
>> Is this what you mean?
>> On Fri, Dec 9, 2011 at 2:00 PM, Thiago Guerrera<thigm85 at gmail.com> wrote:
>>> Hello R users,
>>> I have used the R package BRugs in some applications but one thing that
>>> annoy me is the functions to retrieve samples ("samplesSample") or
>>> statistics ("samplesStats"). They take a really long time to return the
>>> values if the size of the chains in question is moderate, for example
>>> 10.000 samples. Does this happens with every one or there is a better way
>>> to obtain such values?
>>> Thiago Guerrera.
>>> [[alternative HTML version deleted]]
>>> R-help at r-project.org mailing list
>>> PLEASE do read the posting guide
>>> and provide commented, minimal, self-contained, reproducible code.
More information about the R-help