Weekly links round-up: 06/02/2015

Brian McGill posted a thoughtful, in-depth piece on complex regression models, and the use of many terms in such models. The overall conclusion is that inclusion of many terms in a regression model cannot make up for poor experimental design, and that it may in fact create further problems like increased chance of collinearity between predictors and invalidity of p-values.

For those interested in Bayesian statistics, there’s a few links: firstly a nice paper on Bayesian model selection by Hooten and Hobbs; and Petr Keil has posted materials for a Bayesian statistics course, and an introduction to Bayesian Principal Components Analysis developed with Jan Smycka (which they welcome feedback on).

PhD students coming to the end of their programme may be interested in this post from Markus Eichhorn on finishing the PhD thesis and this article from the Guardian on how to successfully apply for a postdoc.

Jobs Postdoc at Sheffield for an evolutionary ecologist which involves modelling herbicide resistance of black-grass. Postdoc at Salford for an ecological/epidemiological modeller.

Advertisements

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