R Weekly 2017 Issue 2

09 Jan 2017

Hello and welcome to this new issue!

Highlight

• Interview with Joe Cheng - the inventor and lead developer for Shiny

• Shiny Chart Builder – Explore your database with a point-and-click interface

• Understanding mixture models and expectation-maximization (using baseball statistics)

Tutorials

• Why do Decision Trees Work?

• Explaining the Almon Distributed Lag Model

• How to Write Pelican Blog Posts using RMarkdown & Knitr

• Text Processing using the textTinyR package

• Cross-validation of topic modelling

• Linear Regression from Scratch in R

• Package evolution – changing stuff in your package

• Easy machine learning pipelines with pipelearner: intro and call for contributors

• truncated normal algorithms

• Use CSS to format markdown or HTML files

• Django and R on Heroku

• Handling Class Imbalance with R and Caret – Caveats when using the AUC

• (lazy)Loading Cached Chunks into an Interactive R Session

• Shiny Chart Builder – Explore your database with a point-and-click interface

• Animations in R using Plotly

• Distributional Semantics in R: Part 2 Entity Recognition w. {openNLP}

• 3PL models viewed through the lens of total probability theorem

• convert graphic object to tree object using treeio

• tidyquant: Bringing Quantitative Financial Analysis to the tidyverse

• What is a Linear Trend, by the way?

• A Theory of Nested Cross Simulation

Insights

• Three reasons to learn R today

• Our R package roundup

• Why R is the best data science language to learn today

• The Most Important Commodity in 2017 is Data

• Simple Features Now on CRAN

R in the Real World

• Explaining Statistical Goodness of fit Tests with Beer (Statistics)

• GooglyPlus: yorkr analyzes IPL players, teams, matches with plots and tables

• Removing Personal Bias From Flu Severity Estimation (a.k.a. Misery Loves Data)

• Thematic Interactive Map - The last month, yep, the past year I’ve working on create maps easily with highcharter. When I saw this chart I took as challege to replicate this nice map in using highcharter.

• Analyzing emotions in video with R

• Inflation of the US Dollar [OC]

• End of Season Fantasy Round Up: Draft Values, Busts, and Best Free Agents of the 2016 Season

• City of Fort Collins Spending Analysis

• Gene homology Part 3 – Visualizing Gene Ontology of Conserved Genes

• An overview of text mining visualisations possibilities with R on the CETA trade agreement

• Understanding mixture models and expectation-maximization (using baseball statistics)

• Forcasting Natural Catastrophes (is rather difficult)

• 10,000 CRAN Packages - The R package system continues to expand as the number of packages on CRAN is about to blow through the 10,000 mark sometime soon.

R in Organization

• Interview with Joe Cheng - the inventor and lead developer for Shiny

• The Flexibility of Remote and Local R Workspaces

• Integrate Power BI Desktop with R

Videos and Podcasts

• NSSD 30 - Philately and Numismatology Hilary and Roger follow up on open data and data sharing in government. They also discuss artificial intelligence, self-driving cars, and doing your taxes in R.

Resources

• Cheatsheet: Data Science for Big Data with Spark and sparklyr

• The biggest R stories from 2016

New Packages and Tools

• ebbr - Introducing the ebbr package for empirical Bayes estimation (using baseball statistics)

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

• Shiny Chart Builder - Shiny app for building charts with a point-and-click interface

• tidyquant - Seamless interaction between quantitative financial analysis packages and the tidyverse.

• TeleR - With a Telegram account connected to @TeleR, the force of R can always be with me.

New Releases

• RcppTOML 0.1.0 - ‘Rcpp’ Bindings to Parser for Tom’s Obvious Markup Language

• jsonlite 1.2 - A Robust, High Performance JSON Parser and Generator for R

• digest 0.6.11 - R package to create compact hash digests of R objects

Jobs

• R Jobs for R users – 8 jobs from around the world (2017-02-03)

R Project Updates

Updates from R Core,

• Header Rconfig.h no longer defines SUPPORT_OPENMP; use _OPENMP as long documented.

• HAVE_UINTPTR_T is now defined where appropriate by Rconfig.h so that it can be included before Rinterface.h when CSTACK_DEFNS is defined and a C compiler (not C++) is in use. Rinterface.h now includes C header stdint.h or C++11 header cstdint where needed.

• No-longer-installed headers GraphicsBase.h, RGraphics.h, Rmodules/RX11.h and Rmodules/Rlapack.h which had a LGPL license no longer do so.

• Enviromental variable _R_CHECK_TESTS_NLINES_ controls how R CMD check reports failing tests (see §8 of the R Internals manual).

• The default method for quantile(x, prob) should now be monotone in prob, even in border cases, see PR#16672.

• The check for the zlib version is now robust to versions longer than 5 characters, including 1.2.10.

• seq("2", "5") now works too, equivalently to "2":"5" and seq.int().

• seq.int(to=1, by=1) is now correct, other cases are integer (instead of double) when seq() is integer too, and the “non-finite” error messages are consistent between seq.default() and seq.int(), no longer mentioning NaN etc.

Call for Participation

• eoda celebrates R – don’t miss the date and win!

• EVENT Announcement: EARL Conferences 2017

• R/Finance 2017: Call for Papers

Upcoming Events

• rstudio::conf 2017 January 12 and 14, 2017
  The conference about all things R and RStudio.
  

• SatRday #2 Conference - Cape Town, South Africa 18 Feb, 2017
  The R community and some of South Africa’s most forward thinking companies have come together to bring satRday to Cape Town. This conference brings an opportunity to hear from and network with top Researchers, Data Scientists and Developers from all over the country and the world.

• useR! 2017 July 4, 2017
  The annual useR! conference is the main meeting of the international R user and developer community. Its program consists of both invited and user-contributed presentations.
  

Quotes of the Week

"#rstats is not a DSL. It’s a language for writing DSLs, which is something altogether more powerful" — @jcheng

— Hadley Wickham (@hadleywickham) January 5, 2017