R resources

 

Here’s a collection of links and resources related to learning and working with R, as well as analysis techniques related to the topic of maximum likelihood analysis. I drafted this list as part of a course I taught on maximum likelihood analysis in the summer of 2012.



Books:


Ecological models and data in R by Ben Bolker. If you could get just one book, get this one. This book is super approachable, and covers pretty much everything we'll touch on in class and more, from the basic ideas right up through sophisticated applications of the techniques to real examples, and includes example code in R. This book and the primary R package that we'll be using for all of our lab work were made to go hand-in hand. The website includes additional code and resources, as well as a (free!) pdf copy of a draft version of the book (some stuff is missing/wrong, but hey, it's free). Otherwise it goes for about $55.00 on Amazon - well worth it in my opinion.


The ecological detective by Hilborn and Mangel. This is a slightly older book than the rest, but the introductory chapters do an awesome job of thinking about what it means to try to model ecological data, and it presents a nice introduction to the concepts of maximum likelihood analysis.


The R book - Michael Crawley. Popular and 'comprehensive' guide to R; I haven't used it before but people seem to like it.


Use R! series - This is an excellent (and growing!) set of short reference books focusing on performing different kinds of analyses specifically in R, including example code and relevant theory. Most or all of them are available through MSU's library system as e-books/pdfs. Particularly appropriate at the introductory level might be "A Beginner's Guide to R" by Alain Zuur et al.


Models for Ecological Data - by Jim Clark. This is a more advanced book, that, in addition to covering concepts associated with maximum likelihood analysis, delves deeply into the application of Bayesian approaches to ecological data. Includes examples in R and WinBUGS (?).



Websites:


R reference card - 4 page pdf briefly listing some important/useful/basic R functions and commands


Ecological models and data in R Wiki - this is where Ben Bolker maintains a list of errata and other resources/answers related to his book


Errata for "The Ecological Detective"


Snippets of R code for "The Ecological Detective"


http://www.inside-r.org/


http://pairach.com/2012/02/26/r-tutorials-from-universities-around-the-world/


Wikipedia - actually has extremely good resources on different kinds of probability distributions and other relevant topics.


MLE introduction


Poisson derivation


RA Fisher and Maximum Likelihood



Useful tools:


The R Project for Statistical Computing


R CRAN - source of packages, etc. for R


R Seek - R search engine; cross references with a bunch of websites and R help lists to look for answers.


R graphics gallery - Want to know how to make a particular kind of plot in R? Browse through the thumbnails of tons of different graphs, all made in R, until you find one that looks right. Then look behind the scenes at the code that created it.


Stack exchange - R help site.


graphical user interfaces (GUIs) for R - The default script editor for windows is rather limited (Mac users, the default editor is probably fine). If you want to explore your options for fancier GUIs for interacting with R, this is the place to start. I might recommend R Commander, RStudio, or Rkward, depending on your platform.


There are a host of blogs, help lists (often for specific R packages/analyses), and other websites out there. One of the great things about working with R is that if you get stuck, it's likely that someone else has had the same problem/error and asked a question/posted an answer on the internet. So, google (or R seek) is your friend when working in R. And, if you don't find an answer to your question, there are places to ask it where experts can try to answer - just pay attention to the posting guidelines so they don't get cranky (which they are sometimes prone to, likely because of the amount of time they spend working on computers).



Recommended R packages:


bbmle - Ben Bolker's maximum likelihood estimation package


emdbook - package of data and auxiliary tools related to Ben Bolker's 'Ecological models and data in R' book.


reshape - great package for reorganizing and formatting data in R; like Excel's pivot table, but awesomer.


lubridate - new R package for dealing with time/date formats, a perennial problem in analyses


ggplot2 - excellent graphics package





Colin Kremer


W. K. Kellogg Biological Station

Michigan State University

3700 E. Gull Lake Dr.

Hickory Corners, MI 49060


kremerco (at) msu (dot) edu