Posts tagged "R"

From Hadley Wickham’s dplyr tutorial at useR! 2014. See this post for more.

From this repo. The network contains almost 9,000 bill cosponsorships between 559 MPs of the ongoing 14th legislature. The size of the nodes represents the weighted degree (the number of ties) of each MP with his cosponsors. The ultra-central Socialist is, unsurprisingly, Bruno Le Roux, the majority leader.

From this repo. The network contains almost 9,000 bill cosponsorships between 559 MPs of the ongoing 14th legislature. The size of the nodes represents the weighted degree (the number of ties) of each MP with his cosponsors. The ultra-central Socialist is, unsurprisingly, Bruno Le Roux, the majority leader.

Legislative cosponsorships from this repo I just finished updating. The data are higher quality (removed a bit of noise, added MP details for session 12), the network plots are more flexible, and the centrality measures are now properly weighted.

The plotting function is ggnet, which should get a light update soon-ish to support additional layout parameters.

Legislative cosponsorships from this repo I just finished updating. The data are higher quality (removed a bit of noise, added MP details for session 12), the network plots are more flexible, and the centrality measures are now properly weighted.

The plotting function is ggnet, which should get a light update soon-ish to support additional layout parameters.

New devtools guarantee: if because of a devtools bug, a CRAN maintainer yells at you, I’ll send you a hand-written apology note. Just forward me the email and your address.
devtools/NEWS.md at master · hadley/devtools
dartthrowingchimp:

Relative risk of new episodes of state-led mass killing in 2013, according to an ensemble of statistical models based on country-year data.
Countries in red had ongoing episodes at the start of the year. Darker grey means higher risk.
For more info on this work and the upcoming early-warning system it will inform, see this APSA 2013 paper of mine.

Method here:

Write an academic paper.
Make it freely available on an open access repo.
Host the extra files (data, appendix etc.) on Google Drive.
If there’s code, host it on GitHub.
Advertise on Tumblr by picture, quote, whatever.
Get actually read by academics of all sorts.
PROFIT
Also, my eye caught that, for some reason:

dartthrowingchimp:

Relative risk of new episodes of state-led mass killing in 2013, according to an ensemble of statistical models based on country-year data.

Countries in red had ongoing episodes at the start of the year. Darker grey means higher risk.

For more info on this work and the upcoming early-warning system it will inform, see this APSA 2013 paper of mine.

Method here:

  1. Write an academic paper.
  2. Make it freely available on an open access repo.
  3. Host the extra files (data, appendix etc.) on Google Drive.
  4. If there’s code, host it on GitHub.
  5. Advertise on Tumblr by picture, quote, whatever.
  6. Get actually read by academics of all sorts.
  7. PROFIT

Also, my eye caught that, for some reason:

wat

Repo updates

Version 4.4 of the course material takes about 600 MB less disk space, thanks to the BFG Repo-Cleaner. GitHub rocks. This amount of control over file versioning is something that I would miss if it were to go away.

I’m also currently digging that repo by Anthony Damico. The GET and POST tricks to automate access to protected online resources is über-cool. I have spent too much time logging into portals, so this is something that I am going to try learning quickly.

I’m also digging this repo at the moment. It contains a very simple example of a makefile that allows to run R completely in the background. The analysis is interesting but it’s taking me some time to catch up with some of the methods.

Two examples of faceted plots:

  1. Party identification as predicted by a bunch of ANES variables through time, i.e. the original example of Gelman and Hill’s "secret weapon" plot.
  2. The 90/10 percentile income ratio in European countries and the United States, as computed from the Luxembourg Income Study. See this Gist on how to get the data through Quandl.

GitHub data

In a smart move, GitHub has added support for CSV and TSV online data exploration:

The course datasets have been renamed accordingly. I also added functions to read compressed datasets.

Sic transit data mundi

I’m working on a package to make Quality of Government data easy to manipulate with R. All comments welcome!