transferring installed packages to a different installation of R

It used to take me a while to reinstall all the R packages that I use after upgrading to a new version of R.  I couldn’t think of another way to do this than to create a list of installed packages by examining the R package directory, and to manually select and install each one of those packages in the new version of R.  In order to ensure that my home and office installation of R had the same packages installed, I did something similar.

I recently discovered that there is a much, much easier way to transfer the packages that you have installed to a different installation of R.  I found some R code on the web that I adapted to my needs.  Here is what you need to do:

1. Run the script “store_packages.R” in your current version of R.

# store_packages.R
#
# stores a list of your currently installed packages

tmp = installed.packages()

installedpackages = as.vector(tmp[is.na(tmp[,"Priority"]), 1])
save(installedpackages, file=”~/Desktop/installed_packages.rda”)

(Make sure that all the quotation marks in the script are straight.  The scripts will generate an error if they include any curly quotation marks.  For some reason, when I saved this blog entry, some quotation marks changed to curly ones.  WordPress is probably to blame for this problem, which I have not been able to fix.)

2. Close R.  Open the installation of R that you want the packages to be installed in.

3. Run the script “restore_packages.R”.

# restore_packages.R
#
# installs each package from the stored list of packages

load(“~/Desktop/installed_packages.rda”)

for (count in 1:length(installedpackages)) install.packages(installedpackages[count])

Note that if you want to install the list of packages in an installation of R on a different computer, you should transfer the .rda file that is created by the store_packages script to that computer, and make sure that the path for the “load” command in the restore_packages script is set to the right location.

Correlation plot matrices using the ellipse library

My new favorite library is the ellipse library. It includes functions for creating ellipses from various objects. It has a function, plotcorr() to create a correlation matrix where each correlation is represented with an ellipse approximating the shape of a bivariate normal distribution with the same correlation. While the function itself works well, I wanted a bit more redundancy in my plots and modified the code. I kept (most of) the main features provided by the function and I’ve included a few: the ability to plot ellipses and correlation values on the same plot, the ability to manipulate what is placed along the diagonal and the rounding behavior of the numbers plotted. Here is an example with some color manipulations. The colors represent the strength and direction of the correlation, -1 to 0 to 1, with University of Rochester approved red to white to blue.

First the function code:

Creating spaghetti plots of eye-tracking data in R

I’ve been working on consolidating all the different R functions I’ve written over the years for plotting my eye-tracking data and creating just one amazing super-function (based on the ggplot2 package) that can do it all. Here’s a first attempt that anybody with the right kind of dataset should be able to use to create plots like the ones below (generated from fake data. The R code that generates the data is included at the end of the post). If you find this code helpful, please consider acknowledging it via the following URL in your paper/presentation to spread the word:

http://hlplab.wordpress.com/2012/02/27/creating-spaghetti-plots-of-eye-tracking-data-in-r/

Left: Empirical means with error bars indicating standard error for four experimental conditions. Contrast presence is coded in color, adjective type in line type. The first vertical line indicates adjective onset, the second ones indicate mean noun onset in each contrast condition. Right: Smoothed model estimates of proportions in each condition, with ribbons indicating 95% confidence intervals. Data from different subjects is plotted in different panels.

R code for LaTeX tables of lmer model effects

Here’s some R code that outputs text on the console that you can copy-paste into a .tex file and creates nice LaTeX tables of fixed effects of lmer models (only works for family=”binomial”). Effects <.05 will appear in bold. The following code produces the table pasted below. It assumes the model mod.all. prednames creates a mapping from predictor names in the model to predictor names you want to appear in the table. Note that for the TeX to work you need to include \usepackage{booktabs} in the preamble.

Using WinBUGS on an PPC OSX laptop connected to a Linux server

While R is an excellent tool for a wide variety of statistical analyses, it’s not the only game in town. Practitioners of Bayesian statistics have a few other tools that complement R nicely. One case where R originally lagged was in offering a general-purpose MCMC sampler. That situation has largely changed, but there are still cases where you might want to look outside of the R toolbox. In particular, certain Bayesian stats books are written with the assumption that exercises and examples can be executed in WinBUGS. While there is (just) another Gibbs sampler that runs natively on OSX and linux, JAGS can’t read WinBUGS .odc files.

Read on to see how I got WinBUGS running on my PowerPC OSX laptop connect to a linux server.

Plotting effects for glmer(, family=”binomial”) models

UPDATE 12/15/10: Bug fix. Thanks to Christian Pietsch.

UPDATE 10/31/10: Some further updates and bug fixes. The code below is the updated one.

UPDATE 05/20/10: I’ve updated the code with a couple of extensions (both linear and binomial models should now work; the plot now uses ggplot2) and minor fixes (the code didn’t work if the model only had one fixed effect predictor).  I also wanted to be clear that the dashed lines in the plots aren’t confidence intervals. They are multiples of the standard error of the effect.

Here’s a new function for plotting the effect of predictors in multilevel logit models fitted in R using lmer() from the lme4 package. It’s based on code by Austin Frank and I also borrowed from Harald Baayen’s plotLMER.fnc() (package languageR). First a cool pic:

Predicted effect of speechrate on complementizer-mentioning

These plots contain the distribution of the predictor (x-axis) against the predicted values (based on the entire model, y-axis) using hexbinplot() from the package hexbin. On top of that, you see the model prediction fo the selected predictor along with confidence intervals. Note that the predictor is given in its original form (here speech rate) although it was entered into the model as the centered log-transformed speechrate. The plot consideres that. Of course, you can configure things.