Issue 25

14 Nov 2016

Hello and welcome to the new issue of R Weekly!

Highlight

Digested by R Weekly Members

• Dot map of the 2016 presidential election in Dallas-Fort Worth - source code on GitHub

• One-page R Survival Guides

News & Blog Posts

R & Election - After

• Not all forecasters got it wrong: Nate Silver does it again (again)

• Dot map of the 2016 presidential election in Dallas-Fort Worth - source code on GitHub

R & Election - Before

• Deceit in Politics; An Analysis of PolitiFact Data

• A Bayesian election prediction, implemented with R and Stan

• 2016 US Presidential Election Forecast

• Analyzing Emotions using Facial Expressions in Video with Microsoft AI and R

• Sentiment Analysis of the Three 2016 Presidential Debates

• Election 2016: Tracking Emotions with R and Python

• The Polls of the Future Are Reproducible and Open Source

• R resources for Election Night

• Election 2016 — Debate Three on Twitter

• State and National Poll Aggregation

Tutorials

• You should re-encode high cardinality categorical variables

• Hierarchical models with RStan (Part 1)

• Laplace noising versus simulated out of sample methods (cross frames)

• Conditional Formatting of a Table in R

• Reproducible Finance with R: The Sharpe Ratio

• Filled Chord Diagram in R using Plotly

• Creating figures like the paper ‘Completeness of Digital Accessible Knowledge of Plants of Ghana’ Part 2

• variance of an exponential order statistics

• sparklyr: a test drive on YARN

• Raccoon Ch 2.2 – 2 sample t-test and paired t

• Image Classification in R using trained TensorFlow models

• Webscraping with rvest: So Easy Even An MBA Can Do It!

• A Simple Guide to S3 Methods

• Timeseries forecasting using extreme gradient boosting

• I’ve started writing a ‘book’: Functional programming and unit testing for data munging with R

R in the Real World

• How the Ghana Floods animation was created

R in Organization

• AirBnB grows by sharing data scientist knowledge

• R Views – a new perspective on R and RStudio

• Data Manipulation with sparklyr on Azure HDInsight

• New Course Series: Applied Finance with R

• A Free Interactive Plotly R Tutorial

R in Academia

• New paper: “SPLASH, a hashed identifier for mass spectra”

• Presenting Weighted Effect Coding

Resources

• 7th MilanoR Meeting: download presentations

• One-page R Survival Guides

New Releases

• rms 5.0-0 - Regression Modeling Strategies.

• anytime 0.1.0 - anytime aims to convert anything in integer, numeric, character, factor, ordered, … format to POSIXct (or Date) objects.

• gettz 0.0.3 - gettz provides a possible fallback in situations where Sys.timezone() fails to determine the system timezone.

• DOM 0.4 - The main change in this version is the addition of new functions that allow control over the Cascading Style Sheet (CSS) content of a web page.

• Shiny Server (Pro) 1.5

• pkgKitten 0.1.4 - Creating R Packages that purr

New Packages & Tools

• USA.county.data - Open dataset of counties from the United States

• knowledge-repo - A next-generation curated knowledge sharing platform for data scientists and other technical professions from AirBnB.

• packagedocs - Build an R Package Documentation Website.

• BelgiumMaps.StatBel - R package with Administrative boundaries of Belgium.

• regtools - Tools for linear, nonlinear and nonparametric regression and classification.

• crul - an HTTP client.

• trump - Instantly segfault your R session.

• ggloop - Create ‘ggplot2’ Plots in a Loop

R Project Updates

Updates from R Core.

• The JIT byte code compiler is now enabled by default at level 3. This means functions will be compiled on first sue and top level loops will be compiled and then run. Thanks to Tomas Kalibera for extensive work to make this possible. For now, the compiler will not compile code containing explicit calls to browser(). This is to support single stepping from the browser() call. The JIT can be disabled using compiler::enableJIT().

• R CMD check --no-stop-on-test-error now continues running tests (in tests/) even if they give errors.

Upcoming Events

• rstudio::conf 2017 January 13 and 14, 2017
  The conference about all things R and RStudio.
  

Quote of the Week

R-devel now has "level 3" byte-compiling, and e.g. "empty" nested loops run about twice as fast.https://t.co/bnNo8XrI6F pic.twitter.com/VFjNbnYZGB

— Dirk Eddelbuettel (@eddelbuettel)

Here's a list of #rstats packages, catered to those doing statistics in sports pic.twitter.com/UqMcrojEoL

— Michael Lopez (@StatsbyLopez)