Prospective students

I am open to new collaborations with students if I think that I can add value to a project. My background is in quantitative conservation ecology, with particular interest in Darwinian fitness metrics, ecological physiology (especially stress physiology), disease ecology, animal movement and behaviour. I help run a remote field research centre, the Mogalakwena Research Centre, where I can host student projects.

I favour Bayesian and Information-theoretic approaches, where applicable, which can often be a barrier for students who are not familiar with the concepts. I usually recommend that students unfamiliar with Bayesian approaches start with a comparison to the basic t test, a Frequentist test which most students learn about during their undergraduate training. From there, another good introduction into the subject of Bayesian inference can be found here.

I work in R, Rstudio, and with Bayesian samplers such as OpenBUGS, JAGS, and STAN. I expect students to learn how to use these tools with my assistance and mentoring (see some notes on the software below). I also like to prepare presentations and manuscripts directly from R using rmarkdown, or using LaTeX, and will expect the same from advanced students.

Please contact me if you would like to collaborate: pnlaver (at) gmail (dot) com

Notes on getting started with R

All of the software listed here is free and open source.

I use R for data management, data analysis, and report or manuscript writing. I recommend its use for all students and collaborators. The coding aspect may initially be daunting, but it comes with great advantages: having a record of your work, having a repeatable workflow for lengthy analysis, and having ‘recipe’ for communicating your work with other people.

What is R? Start here:

https://www.r-project.org/about.html

Download R here:

https://cran.r-project.org/

R can be used on its own, but most people use a graphical user interface (GUI) that brings various aspects of R together: data management, code, plotting of figures. The code is also coloured to make writing and reading code easier. There are several R GUIs available, but I recommend RStudio. Once you have R installed, and then RStudio installed, you will be able to work exclusively in RStudio (but it needs R running in the background).

Download RStudio here:

https://www.rstudio.com/

Basic R tutorial here:

http://tryr.codeschool.com

Optional:

For writing reports and presenting your work, I recommend that you use Rmarkdown, which is a way of writing R scripts that allows you to integrate your code, your analysis, your output, and explanatory text in one document. This is an optional approach, but one that can prove fruitful in the long term.

Start with Rmarkdown here:

http://rmarkdown.rstudio.com/

Rmarkdown mini tutorial here:

http://www.jacolienvanrij.com/Tutorials/tutorialMarkdown.html

Optional:

I often use a Bayesian approach to do my analysis. There are three options for software needed to run Bayesian analyses: OpenBUGS, JAGS, STAN. I recommend STAN.

Start with STAN here:

http://mc-stan.org/ and http://mc-stan.org/users/

You’ll need to interface with R via RStan:

http://mc-stan.org/users/interfaces/rstan.html

and

https://github.com/stan-dev/rstan/wiki/RStan-Getting-Started