Do you remember the notorious hurricane / himmicane study (Jung et al., PNAS, 2014)? At the time, there was a heavy backlash against the study, and probably rightly so, as the statistical analysis turns out to be highly unstable against a change of the regression formula. You can find some links here. Over the years,…
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…
A guest post by @MaxMaPichler, MSc student in the Group for Theoretical Ecology / UR Artificial neural networks, especially deep neural networks and (deep) convolutions neural networks, have become increasingly popular in recent years, dominating most machine learning competitions since the early 2010’s (for reviews about DNN and (D)CNNs see LeCun, Bengio, & Hinton, 2015). In ecology,…
This is a somewhat belated introduction of a package that we published on CRAN at the beginning of the year already, but I hadn’t found the time to blog about this earlier. In the R environment and beyond, a large number of packages exist that estimate posterior distributions via MCMC sampling, either for specific statistical models (e.g.…
As I said before, I firmly side with Andrew Gelman (see e.g. here) in that model checking is dangerously neglected in Bayesian practice. The philosophical criticism against “rejecting” models (double-using data etc. etc.) is all well, but when using Bayesian methods in practice, I see few sensible alternatives to residual checks (both guessing a model and…
I just released a small R package that I have been working on for a while. The motivation for this package came from the observation that I kept on receiving questions about residual checks for GLMMs. The problem that people have is that they have fitted their GLMM, maybe they tested it for overdispersion and…
The central quantity in Bayesian inference, the posterior, can usually not be calculated analytically, but needs to be estimated by numerical integration, which is typically done with a Monte-Carlo algorithm. The three main algorithm classes for doing so are Rejection sampling Markov-Chain Monte Carlo (MCMC) sampling Sequential Monte Carlo (SMC) sampling I have previously given…
I have blogged a few days a ago about a study by Kiju Jung that suggested that implicit bias leads people to underestimate the danger of female-named hurricanes. The study used historical data to demonstrate a correlation between femininity and death-toll, and subsequent experiments seemed to show that people indeed estimate hurricanes to be less…
Approximate Bayesian Computation (ABC) is an umbrella term for a class of algorithms and ideas that allow performing an approximate estimation of the likelihood / posterior for stochastic simulation models when the likelihood cannot be explicitly calculated (intractable likelihood). To give you the idea in a nutshell: to approximate the likelihood, consider that for a…
A colleague had a question about sampling design and we didn’t find a good answer … so, if you like to solve riddles, you might like that one: We want to distribute n=3 plant species across k=12 x m=12 grid cells, in a way that no individual has another individual of it’s own species in…
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”…
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…
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…
theoretical ecology 