tl;dr: DHARMa tests will pick up on overdispersion before you see a rise of Type I error. Overdispersion is a common problem in GL(M)Ms with fixed dispersion, such as Poisson or binomial GLMs. Here an explanation from the DHARMa vignette: GL(M)Ms often display over/underdispersion, which means that residual variance is larger/smaller than expected under the…
This (co-)guest post by Carsten F. Dormann & Florian Hartig summarizes a comprehensive review on model averaging for predictive inference, just published in Ecological Monographs.
A comment on a recent post gave me the motivation to try out the new EasyABC package for R, developed by Franck Jabot, Thierry Faure, Nicolas Dumoulin and maintained by Nicolas Dumoulin. Approximate Bayesian Computation (ABC) is a relatively new method that allows treating any stochastic model (IBM, stochastic population model, …) in a statistical…
Approximate Bayesian Computing and similar techniques, which are based on calculating approximate likelihood values based on samples from a stochastic simulation model, have attracted a lot of attention in the last years, owing to their promise to provide a general statistical technique for stochastic processes of any complexity, without the limitations that apply to “traditional”…
In the last while, a number of people that were fighting with slow R code asked me for advice. Commendably, they had ticked off the usual suspects such as Do you loop over things that could be vectorized? Do you have everything as arrays, do you avoid resizing, do you avoid accessing with $NAMES? Have…
In a recent post, I reported on the scaling of data transfer rates in the RNetLogo interface between R and NetLogo. I had already updated this post to include some preliminary fixes that were communicated to me by Jan Thiele, who maintains the RNetLogo package, but there was the still the information that Jan was…
IMPORTANT UPDATE 16.4.: Since this was posted, Jan Thiele made some improvements to RNetLogo, and there will be a new version out soon with improved performance for large data transfers. Jan also noted that the new NetLogo version 5.0 is considerably faster than the 4.x line which was used to create the figures below, so…
Last week, I gave a seminar about MCMC chain analysis and convergence diagnostics with coda in R, and I thought a summary would make a nice post. Note: this post is about checking the convergence of the MCMC – a more recent post explains how to check the adequacy of model assumptions in a Bayesian…
NOTE: these instructions are outdated, an update has been posted here. WinBugs and its open source counterpart OpenBugs are the most commonly used software packages for evaluating the posterior distribution of a Bayesian model by means of a Markov Chain Monte Carlo algorithm (see e.g. my previous post for more details on MCMC algorithms). WinBugs…
The new Eclipse release 3.7 (Indigo) is out since a few weeks. Contrary to a few remarks that I read on the web, it worked fine for me to simply add http://download.eclipse.org/releases/indigo to my update sites to upgrade from 3.6 to 3.7 without loosing any settings or packages.
The image was created with an example from the scipy homepage.
I was looking for a journal bibstyle today, and I found an excellent solution at the LaTeX Bibliography Styles Database. The maintainers state that the database contains style files for currently 2300 journals. Great website!
While there are certainly good software packages out there to do the job for you, notably BUGS or JAGS, but also our own BayesianTools package with general-purpose MCMC samplers, it is instructive to program a simple MCMC yourself. In this post, I give an educational example of the Bayesian equivalent of a linear regression, sampled…
Most scientists have never been formally educated in computer science, yet, a ever bigger chunk of our daily work is in fact the work of a computer scientist – we organize and process large amounts of data, develop algorithms and write smaller or larger pieces of software and scripts. Eventually, everyone finds a personal solution…
I was looking for a open source hosting service today. So far, I have been with sourceforge, but I thought I should give google code’s project hosting a go. To cut a long story short, I wasn’t very pleased with it. Initially, I thought the pros and cons are more or less: Sourceforge: + largest…
I have been testing the StatET Eclipse plug-in for R for a few weeks now, and I am very pleased with it. I guess it depends on whether you like eclipse, but if you do, I think it’s a great tool. The only major feature I was missing is a better integration of the R…
Detexify translates handwritten (or rather mouse-drawn) symbols to latex code … nice idea guys !!!
At first I thought I’d encountered the bike of a true statistician, but it was a “RAGAZZI”, apparently a brand that is sold through a German discounter. See below
When having to use Microsoft Word, I’m using the very nice bibtex4word plugin since a while, which is provided by Mike Brookes. It provides a simple possibility for creating references from a bibtex library in word, including (limited) possibilities to use LaTeX bibstyles. All details can be found here http://www.ee.ic.ac.uk/hp/staff/dmb/perl/index.html
theoretical ecology 