Alp Kucukelbir, Rajesh Ranganath, Dave Blei, and I write: We describe an automatic variational inference method for approximating the posterior of differentiable probability models. Automatic means that the statistician only needs to define a model; the method forms a variational approximation, computes gradients using automatic differentiation and approximates expectations via Monte Carlo integration. Stochastic gradient […]

**Statistical computing**category.

## Stan Down Under

I (Bob, not Andrew) am in Australia until April 30. I’ll be giving some Stan-related and some data annotation talks, several of which have yet to be concretely scheduled. I’ll keep this page updated with what I’ll be up to. All of the talks other than summer school will be open to the public (the […]

## This has nothing to do with the Super Bowl

Joshua Vogelstein writes: The Open Connectome Project at Johns Hopkins University invites outstanding candidates to apply for a postdoctoral or assistant research scientist position in the area of statistical machine learning for big brain imaging data. Our workflow is tightly vertically integrated, ranging from raw data to theory to answering neuroscience questions and back again. […]

## Six quick tips to improve your regression modeling

It’s Appendix A of ARM: A.1. Fit many models Think of a series of models, starting with the too-simple and continuing through to the hopelessly messy. Generally it’s a good idea to start simple. Or start complex if you’d like, but prepare to quickly drop things out and move to the simpler model to help […]

## Github cheat sheet

Mike Betancourt pointed us to this page. Maybe it will be useful to you too.

## Lewis Richardson, father of numerical weather prediction and of fractals

Lee Sechrest writes: If you get a chance, Wiki this guy: I [Sechrest] did and was gratifyingly reminded that I read some bits of his work in graduate school 60 years ago. Specifically, about his math models for predicting wars and his work on fractals to arrive at better estimates of the lengths of common […]

## Stan comes through . . . again!

Erikson Kaszubowski writes in: I missed your call for Stan research stories, but the recent post about stranded dolphins mentioned it again. When I read about the Crowdstorming project in your blog, I thought it would be a good project to apply my recent studies in Bayesian modeling. The project coordinators shared a big dataset […]

## Expectation propagation as a way of life

Aki Vehtari, Pasi Jylänki, Christian Robert, Nicolas Chopin, John Cunningham, and I write: We revisit expectation propagation (EP) as a prototype for scalable algorithms that partition big datasets into many parts and analyze each part in parallel to perform inference of shared parameters. The algorithm should be particularly efficient for hierarchical models, for which the […]

## Next Generation Political Campaign Platform?

[This post is by David K. Park] I’ve been imagining the next generation political campaign platform. If I were to build it, the platform would have five components: Data Collection, Sanitization, Storage, Streaming and Ingestion: This area will focus on the identification and development of the tools necessary to acquire the correct data sets for […]

*Bayesian Cognitive Modeling* Models Ported to Stan

Hats off for Martin Šmíra, who has finished porting the models from Michael Lee and Eric-Jan Wagenmakers’ book Bayesian Cognitive Modeling to Stan. Here they are: Bayesian Cognitive Modeling: Stan Example Models Martin managed to port 54 of the 57 models in the book and verified that the Stan code got the same answers as […]

## Soil Scientists Seeking Super Model

I (Bob) spent last weekend at Biosphere 2, collaborating with soil carbon biogeochemists on a “super model.” Model combination and expansion The biogeochemists (three sciences in one!) have developed hundreds of competing models and the goal of the workshop was to kick off some projects on putting some of them together intos wholes that are […]

## Stan hits bigtime

First Wikipedia, then the Times (featuring Yair Ghitza), now Slashdot (featuring Allen “PyStan” Riddell). Just get us on Gawker and we’ll have achieved total media saturation. Next step, backlash. Has Stan jumped the shark? Etc. (We’d love to have a “jump the shark” MCMC algorithm but I don’t know if or when we’ll get there. […]

## Just imagine if Ed Wegman got his hands on this program—it could do wonders for his research productivity!

Brendan Nyhan writes: I’d love to see you put some data in here that you know well and evaluate how the site handles it. The webpage in question says: Upload a data set, and the automatic statistician will attempt to describe the final column of your data in terms of the rest of the data. […]

## “The Firth bias correction, penalization, and weakly informative priors: A case for log-F priors in logistic and related regressions”

Sander Greenland sent me this paper that he wrote with Mohammad Ali Mansournia, which discusses possible penalty functions for penalized maximum likelihood or, equivalently, possible prior distributions for Bayesian posterior mode estimation, in the context of logistic regression. Greenland and Mansournia write: We consider some questions that arise when considering alternative penalties . . . […]

## I love it when I can respond to a question with a single link

Shira writes: This came up from trying to help a colleague of mine at Human Rights Watch. He has several completely observed variables X, and a variable with 29% missing, Y. He wants a histogram (and other descriptive statistics) of a “filled in” Y. He can regress Y on X, and impute missing Y’s from […]

## No, Michael Jordan didn’t say that!

The names are changed, but the song remains the same. First verse. There’s an article by a journalist, The odds, continually updated, by F.D. Flam in the NY Times to which Andrew responded in blog form, No, I didn’t say that, by Andrew Gelman, on this blog. Second verse. There’s an article by a journalist, […]

## Stan 2.5, now with MATLAB, Julia, and ODEs

As usual, you can find everything on the Stan Home Page. Drop us a line on the stan-users group if you have problems with installs or questions about Stan or coding particular models. New Interfaces We’d like to welcome two new interfaces: MatlabStan by Brian Lau, and Stan.jl (for Julia) by Rob Goedman. The new […]

## Statistical Communication and Graphics Manifesto

Statistical communication includes graphing data and fitted models, programming, writing for specialized and general audiences, lecturing, working with students, and combining words and pictures in different ways. The common theme of all these interactions is that we need to consider our statistical tools in the context of our goals. Communication is not just about conveying […]

## My course on Statistical Communication and Graphics

We will study and practice many different aspects of statistical communication, including graphing data and fitted models, programming in Rrrrrrrr, writing for specialized and general audiences, lecturing, working with students and colleagues, and combining words and pictures in different ways. You learn by writing an entry in your statistics diary every day. You learn by […]

## Some general principles of Bayesian data analysis, arising from a Stan analysis of John Lee Anderson’s height

God is in every leaf of every tree. The leaf in question today is the height of journalist and Twitter aficionado Jon Lee Anderson, a man who got some attention a couple years ago after disparaging some dude for having too high a tweets-to-followers ratio. Anderson called the other guy a “little twerp” which made […]