Bob Carpenter, Daniel Lee, and I are giving a 3-day short course in two weeks.
Before class everyone should install R, RStudio and RStan on their computers. If problems occur please join the stan-users group and post any questions. It’s important that all participants get Stan running and bring their laptops to the course.
Class structure and example topics for the three days:
Sunday, July 19: Introduction to Bayes and Stan
Morning:
Intro to Bayes
Intro to Stan
The statistical crisis in science
Afternoon:
Stan by example
Components of a Stan program
Little data: how traditional statistical ideas remain relevant in a big data world
Monday, July 20: Computation, Monte Carlo and Applied Modeling
Morning:
Computation with Monte Carlo Methods
Debugging in Stan
Generalizing from sample to population
Afternoon:
Multilevel regression and generalized linear models
Computation and Inference in Stan
Why we don’t (usually) have to worry about multiple comparisons
Tuesday, July 21: Advanced Stan and Big Data
Morning:
Vectors, matrices, and transformations
Mixture models and complex data structures in Stan
Hierarchical modeling and prior information
Afternoon:
Bayesian computation for big data
Advanced Stan programming
Open problems in Bayesian data analysis
Specific topics on Bayesian inference and computation include, but are not limited to:
Bayesian inference and prediction
Naive Bayes, supervised, and unsupervised classification
Overview of Monte Carlo methods
Convergence and effective sample size
Hamiltonian Monte Carlo and the no-U-turn sampler
Continuous and discrete-data regression models
Mixture models
Measurement-error and item-response models
Specific topics on Stan include, but are not limited to:
Reproducible research
Probabilistic programming
Stan syntax and programming
Optimization
Warmup, adaptation, and convergence
Identifiability and problematic posteriors
Handling missing data
Ragged and sparse data structures
Gaussian processes
Again, information on the course is here.
The course is organized by Lander Analytics.
This looks great. Any plans to bring it out west?
Definitely. This is hopefully the first of many of these courses around the world.
We’re hoping to learn a lot doing this one with all three of us in attendance. And to use it as a springboard for the Stan book on applied Bayesian modeling. And, of course, it helps fund the whole Stan enterprise.
When is the european tour?
When anyone asks? My jokes aren’t as good as Andrew’s, but my animations are pretty good…
Is there room on the bus for groupies?
http://www.merriam-webster.com/dictionary/groupie
I strenuously object to the narrow definition offered by Wikipedia.
Great idea but it’s very pricey. Why not produce a short ed x / coursera style video tutorial instead? You’d reach a much larger audience.
Anon:
It’s pricey because it’s a way for us to raise money to continue to support the Stan team. Stan is free and open-source so we have to continue to work on models for funding, and we hope that one motivation for people to pay for the course (in addition to its immediate and direct value for them) is to help support this project.
But, yes, we’re open to other teaching options as well.
Oh, snap. For that kind of money I could have my own Frontiers Research Topic article. Then I wouldn’t even need to learn Stan – it’s like cutting out the middleman!
http://statmodeling.stat.columbia.edu/2013/07/02/19268/
Jrc:
An article in Psych Science is even cheaper, it’s just the cost of 100 people on MTurk and a couple hours computing p-values!
If you add a donate button on mc-stan.org, I will donate money. If you have 1000 users, and each donates something every year (I would happily do that), you should have at least some amount of money to keep going. That way you can teach courses that anyone can attend for free, and don’t have to set up expensive paywalled courses. It might be an experiment worth trying, to just add a donate button.
These kinds of payed courses are really common and there’s no reason we shouldn’t be offering our expertise as well! But it’s important to recognize that we’re not doing payed courses in exclusion to other teaching ventures — we’re regularly giving open and free talks and tutorials, hosting meet ups, and the like all around the world.
I don’t mean it as an either-or proposition. I’m just suggesting that many would probably be happy to donate to Stan.
Patreon account for Stan? Donation per release?
Only half kidding, it might work.
Just as a show of support, I fully approve of you guys selling tickets, much better to sell your knowledge and time (which can’t be copied essentially free of cost) than to sell the software (which can).
Agreed, especially when the group does so much free teaching anyway.