Author Archives: Daniel

Stan workshops at UCLA (6/23) and UCI (6/24)

Posted on by
Reply

This post is by Daniel.

While Bob travels to Boston-ish, I’ll be giving two Stan workshops in Southern California. I’m excited to be back on the west coast for a few days — I grew up not too far away. Both workshops are open, but space is limited. Follow the links for registration.

The workshops will cover similar topics. I’m going to focus more on Stan usage and less on MCMC. If you’re attending, please install RStan 2.6.0 before the workshop.

 

P.S. Congrats, Dub Nation.

Upcoming Stan-related talks

Posted on by
5

If you’re in NYC or Sidney, there are some Stan-related talks in the next few weeks.

 

New York

 

Sydney

  • 4 March. Bob Carpenter: The Benefits of a Probabilistic Model of Data Annotation. Macquarie Uni Computer Science Dept.
  • 10 March, 2–3 PM. Bob Carpenter: Stan: Bayesian Inference Made Easy. Macquarie Uni Statistics Dept. Building E4A, room 523.
  • 11 March, 6 PM, Mithcell Theatre, Level 1 at SMSA (Sydney Mechanics’ School of Arts). Bob Carpenter: RStan: Bayesian Inference Made Easy. Register Now: Sydney Users of R Forum (SURF) (Meetup)

 

 

Stan hack session at Columbia on Saturday

Posted on by
1

[this post is by Daniel]

For those of you in NYC this Saturday, we’re having a Stan hack session from 11 am – 5 pm. A lot of the Stan developers will be around. It’s free, but registration required. See link below.

Bring a laptop, some data, and a model you want to fit. Or come through and work with some other people.

 

Registration link (through the Stan NYC Users meetup group): http://www.meetup.com/Stan-Users-NYC/events/218773576/

 

Stan NYC Meetup – Thurs, July 31

Posted on by
Reply

The next Stan NYC meetup is happening on Thursday, July 31, at 7 pm. If you’re interested, registration is required and closes on Wednesday night: http://www.meetup.com/Stan-Users-NYC/events/193685802/

 

The third session will focus on using the Stan language. If you’re bringing a laptop, please come with RStan, PyStan, or CmdStan already installed.

 

We’re going to focus less on the math and more on the usage of the Stan language. We’ll cover:

• Stan language blocks

• Data types

• Sampling statements

• Vectorization

 

Stan users meetup next week

Posted on by
Reply

We have a Stan users meetup for NYC. We’ll have monthly sessions where we can discuss modeling, success stories, pain points, and really have a chance for the user base and the developers to interact in NYC.

The first meetup will be on Tuesday, 5/13. I’ll be giving a overview of Stan aimed at a general audience. If you’re interested, please register for the group / talk. Space is limited.

 

CmdStan, RStan, PyStan v2.2.0

Posted on by
The Stan Development Team is happy to announce CmdStan, RStan, and PyStan v2.2.0. As usual, more info is available on the

This is a minor release with a mix of bug fixes and features. For a full list of changes, please see the v2.2.0 milestone on stan-dev/stan’s issue tracker. Some of the bug fixes and issues are listed below.

Bug Fixes

  • increment_log_prob is now vectorized and compiles with vector arguments
  • multinomial random number generator used the wrong size for the return value
  • fixed memory leaks in auto-diff implementation
  • variables can start with the prefix ‘inf’
  • fixed parameter output order for arrays when using optimization
  • RStan compatibility issue with latest Rcpp 0.11.0

Features

  • suppress command line output with refresh <= 0
  • added 1 to treedepth to match usual definition of treedepth
  • added distance, squared_distance, diag_pre_multiply, diag_pre_multiply to Stan modeling lnaguage
  • added a ‘fixed_param’ sampler for use with the generated quantities block

For more information and download links, visit Stan’s webpage: http://mc-stan.org
 
— Stan Development Team

(R/Py/Cmd)Stan 2.1.0

Posted on by

We’re happy to announce the release of Stan C++, CmdStan,
RStan, and PyStan 2.1.0.  This is a minor feature release,
but it is also an important bug fix release.  As always, the
place to start is the (all new) Stan web pages:

http://mc-stan.org

 

Major Bug in 2.0.0, 2.0.1

Stan 2.0.0 and Stan 2.0.1 introduced a bug in the implementation
of the NUTS criterion that led to poor tail exploration and
thus biased the posterior uncertainty downward.  There was no
bug in NUTS in Stan 1.3 or earlier, and 2.1 has been extensively tested
and tests put in place so this problem will not recur.

If you are using Stan 2.0.0 or 2.0.1, you should switch to 2.1.0 as
soon as possible and rerun any models you care about.

 

New Target Acceptance Rate Default for Stan 2.1.0

Another big change aimed at reducing posterior estimation bias
was an increase in the target acceptance rate during adaptation
from 0.65 to 0.80.  The bad news is that iterations will take
around 50% longer because of the reduced step size required to
achieve the higher acceptance rate.  The good news is that chains
should be less variable and tails of hierarchical models should
be explored more efficiently (in particular, fewer “stuck” chains).

There are also new configuration parameters that let you control how
long the various phases of adaptation last, with explanations in the
command section of the manual.

The change was motivated by:

    M. J. Betancourt, Mark Girolami.  2013.
    Hamiltonian Monte Carlo for Hierarchical Models.

 

New Features and Minor Bug Fixes

The full list can be found in the release notes at:

    https://github.com/stan-dev/stan/releases/tag/v2.1.0

The highlights in addition to adaptation are a switch to BFGS
as the default optimizer (about an order of magnitude faster and
also more stable), fixes to bugs in bin/print for CmdStan for
matrices, bug fixes to multi_normal_prec(), bernoulli(), bernoulli_logit(),
and Phi(), sd(), var(), all of which are now more stable for vectorization and
arithmetic (former bugs caused crashes, not wrong answers).  Windows
now builds in 64-bit mode, and g++ 4.4 now works.  There were also many
updates to clarify the manual.

As always, we’d love to hear from you on our mailing lists if
you have suggestions, bug reports, problems installing, etc.

 

– Stan Development Team