rStan

classic Classic list List threaded Threaded
3 messages Options
Ian
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

rStan

Ian
The STAN development team just released the first stable version of their Hamiltonian-like MCMC sampler. http://andrewgelman.com/2012/08/a-stan-is-born/

From my reading Hamiltonian MCMC is a much more efficient way of sampling from the posterior than gibbs or metropolis-hastings. It's also super fast: http://andrewgelman.com/2012/08/stan-is-fast/

So can we expect the new StanR compatible book programs shortly?  :)
In all seriousness, they put together a really nice pdf showing how to port worked out bugs examples into STAN.

http://mc-stan.org/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: rStan

John K. Kruschke
Administrator
 
Yes, STAN looks very interesting! Thanks for posting.

On Fri, Aug 31, 2012 at 10:31 AM, Ian [via Doing Bayesian Data Analysis] <[hidden email]> wrote:
The STAN development team just released the first stable version of their Hamiltonian-like MCMC sampler. http://andrewgelman.com/2012/08/a-stan-is-born/

From my reading Hamiltonian MCMC is a much more efficient way of sampling from the posterior than gibbs or metropolis-hastings. It's also super fast: http://andrewgelman.com/2012/08/stan-is-fast/

So can we expect the new StanR compatible book programs shortly?  :)
In all seriousness, they put together a really nice pdf showing how to port worked out bugs examples into STAN.

http://mc-stan.org/


If you reply to this email, your message will be added to the discussion below:
http://doing-bayesian-data-analysis.12272.n6.nabble.com/rStan-tp5000503.html
To start a new topic under Doing Bayesian Data Analysis, email [hidden email]
To unsubscribe from Doing Bayesian Data Analysis, click here.
NAML

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: rStan

Wayne
In reply to this post by Ian
Yes, STAN is amazing. I've used it for a couple of things, and am working my way through DBDA reproducing BUGS results in STAN.

To be honest, I also like the fact that STAN code is executable, while BUGS is a model description. It's a subtle distinction, but it makes a big difference to me.
Loading...