Skip to Content

Blogs

One last Friday post before I sign off. I've seen a few koalas in the time I've spent in Australia, and I expect I'll see a few more on this trip (we'll be visiting Kangaroo Island, the "Madagascar of Australia", which has a healthy koala population). But I've never seen a koala do this:

 

See you in a couple of weeks.

I'm heading out on vacation to my Australian homeland for the next two weeks, but fear not reader friends: we have an awesome lineup of guest bloggers to fill in while I'm away. I don't want to ruin the surprise, but you'll be seeing posts from: experts in data visualization about how they use R; members of the R community on new R packages; R users in industry; and a couple of other surprises.

The social crowdfunding site Kickstarter announced today that it has enabled, via community contributions, $50M in funding in 2012 for new indie games. The second largest category of funding was for independent films. The blog post announcing the news includes analysis (using R, natch) of the breakdown in categories and funding sources. 

In case you missed them, here are some articles from June of particular interest to R users.

RStan is a new package for Bayesian modeling with R. It's faster and can fit more highly-correlated models than the MCMC sampler of BUGS and JAGS.

An update to the ggplot2 "grammar of graphics" package for R is now available on CRAN. This version introduces "Themes" for ggplot2 charts, and makes it possible to define and re-use your own preferences for title fonts and sizes, tick marks, grid color, etc. The system is heirarchically defined, so for example you can choose a font and have it affect just the main title, or every font in the chart.

It all started off as a simple question from Scott Chamberlain on Twitter:

Apparently I'm late to the table with with Psy's Gangnam Style, but now that I've seen it I can't get it this infectious Korea-pop out of my head:

 

Makes me want to visit Seoul again (and Andrew Sullivan has more on the local nods). I'm sure I'll be listening to this more over the long weekend (while the blog takes a break on Monday). Hope you enjoy yours!

For the last decade or so, the go-to software for Bayesian statisticians has been BUGS (and later the open-source incarnation, OpenBugs, or JAGS).

It's a wonderful thing when people make interesting data sets available to the public. When Thomas Jones wrote a paper in Econometrics about the growth of US retail giant Walmart, he made the data he collected about every Walmart store opening in history (location and date) available to the public. Since then, several people have used different techniques to visualize the data. 

You've probably heard (or seen in TV shows) how the unique pattern of rifling in a gunbarrel generates forensic evidence: microscopic scoring on the bullets left at the scene of the crime can be linked to the shooter by matching the marks to the firearm.