Improve Performance
- Windows: Tuneup Utilities
- Mac OSX: Advice
Programing and Data Analysis
- Shell/Terminal- one of the first steps in programming is to get comfortable with command line interface
- Python
- Learn Python the Hard Way – good online book
- The easiest oob python installation for scientific computing (even though Macs come with a python version installed) is Enthought Canopy (EPD). I had some trouble with multiple versions of python and versions of modules but the Enthought version worked like a charm with all the necessary modules and a decent IDE, plus it plays well with IPython and IPython Notebook. Additional (non-Enthought) installation help here (you can use direct Python distribution, MacPorts, Homebrew, or compile from source)
- ipython – the ipython notebook is fantastic!
- Code Academy
- Software Carpentry: Python Lessons
- R
- colClasses to speed data import
- Quick-R (dealing with dates) – one of my favorite references for general R info
- Google Style Guide for R
- Understanding How R Works (environments, packages, objects, etc.)
- R-bloggers – great info on all things R.
- Bayesian R packages – list of R packages using Bayesian computation
- High Performance Computing with R
- C/C++
- Git
- GitHub
- Pro Git (free book)
- Try Git (great intro where you work along on your web browser)
- LifeHacker GitHub intro
- Subversion Control (SVN)
- Regular Expressions
- Make (program that automates workflows)
Data Management
- Software Carpentry: Data Management
- Have a Plan: DataOne
- Microsoft Access
- OpenOffice Base
- Filemaker Pro
- SQL
- Learn SQL the Hard Way – online book
- Metadata
Generalized Linear Mixed Models
R Packages:
- lme4
- glmmPQL {MASS}
- glmmML
- glmmADMB {ADMB} (I have trouble installing properly – haven’t got it to run)
- MCMCglmm
- unmarked (calculate abundance, occupancy, and detection from repeated survey data)
- RStan
- nlme
- geepack
- QICpack
- AICcmodavg
- ggplot2
- lattice
- parallel
- rjags
- R2jags
- R2WinBUGS
- MASS
- IPMpack
- gamm4
- stargazer – The
stargazer
command produces LaTeX code for well-formatted tables that hold regression analysis results from several models side-by-side, as well as summary statistics. More info here - boot
- car – Companion to Applied Regression book by John Fox. Easiest way to run ANOVAs in R.
- knitr – fantastic package for creating markdown files in R. Great with RStudio
Model Selection and Overfitting:
Overfitting (John White Blog)
Useful R Scripts
- Launching multiple iButtons (and here)
- Calculating QIC for Generalized Estimating Equations (GEE) in R
- Visually weighted regression plots
- Accessing and Preparing precipitation data in R from US Historic Climate Network
Computing Resources
- Stack Overflow – Great forum for computer coding (any language including R)
- Cross Validated – Stack Exchange forum for statistics
- Parallel Processing to speed up complex models
- Dropbox – I love Dropbox! I haven’t used my USB flash drive in 2 years since starting with it. I had my dissertation and data backed up via Dropbox and could access it on any computer. 2 GB free. You can also share folders with friends that have Dropbox. I share research files with collaborators. Much better than trying to email files back and forth.
Recommendations for Analysis and Data Presentation
Avoid Dynamite Plots: From Bolker, Resource with some R code, Vanderbilt Wiki (definitely view linked poster)
Other Resources
- Advice on using Markdown instead of MS Word and more here
- How to toggle hidden files on Mac (useful for viewing R and Git files like .RData, .gitnore, .history, .log, etc.)
- Open Source software for Mac (many cross platform options): http://opensourcemac.org/