Posterior Samples

Free download of Hastie, Tibshirani and Friedman’s “Elements of Statistical Learning”. Hastie and Tibshirani did a lot of the early work on Generalised Additive Models and non-parametric smoothing, albeit in a non-Bayesian framework, so this will no doubt be an invaluable resource.

NASA have released a neat visualisation of spatio-temporal deviation from the mean temperature going back to the 1880s. You have to be careful when generating “pretty” graphics, though, because “pretty” infographics can be quite ugly and hard to decode when you’re not taking into account what the data actually represents. Roger Peng reviews Winston Chang’s “R Graphics Cookbook”. The conclusion? A pretty good collection of examples about how to complete particular graphing tasks using ggplot2. It’s a simpler book than Hadley Wickham’s “ggplot2″ and would be a good introduction to using ggplot2 to achieve a particular goal.

JSTOR’s “Register & Read” is in beta now. So if you’re not attached to a university or other institution with a JSTOR subscription, you now have the opportunity to read JSTOR articles for free (but it’s not unlimited, at least not yet). Speaking of free access to published research, this opinion article makes the point that our job as scientists is to contribute to the body of human knowledge for the benefit of society and so if a paper isn’t freely available to the public then it isn’t really “published”. Not only that, but to hide research behind a paywall is immoral. I found it very interesting even if I thought some of his arguments were a bit dismissive.

About these ads

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s