Hello!
Thanks to all for responses. Here is a summary of all answers.
Alun Bedding:
--------------------------------------------------------------
I have run an example recently with 1016 observations. The model was also
very complex but Winbugs dealt with it very well.
Madhuchhanda Bhattacharjee:
--------------------------------------------------------------
I regularly use WinBUGS for analysing high-dimensional bioinformatics
data. For many of these applications not only data size could be in
millions but also "parameters"/variables of interests could be in
thousands. Such models compile and run in WinBUGS without much difficulty.
However, irrespective of PC specification WinBUGS is unable to allocate
memory more than (about) 500 MB and the simulation stops. 200-300 MB
memory requirement to load data in bioinformatics is common and with
largish number of samples to be monitored memory requirement easily
exceeds 500 MB. I had reported this problem to the developers of this
software but unfortunately haven't received any solution so far.
Gentry White:
--------------------------------------------------------------
I have run a spatial model for 115 counties in Missouri, for 4 age groups
and 11 time periods, which is 5060 data points, and more that in terms of
paramters estimated. It takes some time to run but has no problems that I
detected.
Masoud Karimlo:
--------------------------------------------------------------
I tried with 10000 dataset without any problem,very
quick and in a short time, it will be possible to try
with sample size,up to 100000. I think the sample size
is not very important,as you mentioned, the only
limitation is size of the computer.
David Spiegelhalter:
--------------------------------------------------------------
This is difficult to say: the problem is not just BUGS but any MCMC method
will be slow with huge datasets. Completely as a guess, I would be very
cautious if number of observations * number of parameters > 100,000
>
>>Hello!
>>
>>I am starting with Bayesian stats and I did a brief search of
>>available software. BUGS is afcourse at the top of refered and
>>adviced programs. I come from animal breeding field and we have
>>quite large datasets. So I am interested in how many data can
>>one analyze with BUGS. I read FAQ and saw at the end that the
>>only restriction is the size of your computer. However, I would
>>still like to hear of some real life examples. I would also
>>like to hear along the size of the problem i.e. number of records
>>the computer properties.
>>
>>Thank you very much!
>>
>>--
>>Lep pozdrav / With regards,
>> Gregor GORJANC
>>
>>---------------------------------------------------------------
>>University of Ljubljana
>>Biotechnical Faculty URI: http://www.bfro.uni-lj.si
>>Zootechnical Department email: gregor.gorjanc <at> bfro.uni-lj.si
>>Groblje 3 tel: +386 (0)1 72 17 861
>>SI-1230 Domzale fax: +386 (0)1 72 17 888
>>Slovenia
>>---------------------------------------------------------------
>>
-------------------------------------------------------------------
This list is for discussion of modelling issues and the BUGS software.
For help with crashes and error messages, first mail [log in to unmask]
To mail the BUGS list, mail to [log in to unmask]
Before mailing, please check the archive at www.jiscmail.ac.uk/lists/bugs.html
Please do not mail attachments to the list.
To leave the BUGS list, send LEAVE BUGS to [log in to unmask]
If this fails, mail [log in to unmask], NOT the whole list
|