Mastering Data Analysis with R

Sorry about the noisy post title, but it happens to be the name of the book I was working on in the past year, which has been just published at Packt:


Although I do not think that reading this ~400 page book will turn everyone into a true master of R and data analysis, but I believe it can get you on the way. I wrote this book for a relatively large target audience in mind with some prior R experience (like at an introductory university course or MOOC covering how to install R, load CSV files or generate a histogram), but without the time/need to walk through a complete series of books on the stats background, algorithms and domain specific knowledge on handling different data types.

So this is not a reference book, it does not even include a piece of formal mathematical formula, but instead it does provide a practical introduction, many references and hands-on examples on the following topics:

• Reading data from larger text files and databases in an optimal way
• Loading data from the Web via parsing HTML, XML, JSON and interacting with APIs
• Filtering, summarizing and restructuring data
• Building and interpreting generalized linear models
• Traditional multivariate statistical methods for dimension reduction and latent variables
• Classification and clustering, including supervised and unsupervised statistical and machine learning methods
• Handling outliers and missing values
• Processing unstructured text data
• A bit of social network analysis
• Smoothing, seasonal decomposition and modeling time-series 
• Visualizing spatial data

And a free chapter (available from Packt) on “Analyzing the R community”, which combines quite a few techniques described in the above mentioned chapters into an actual use case, including some reproducible examples from some of my past researches on this topic:

• The number of R Foundation members and R conference attendees (previously presented at the useR! 2014 and 2015 conferences besides an interactive webapp on R-activity around the world)
• The number of packages per R package maintainers
• The volume and timeline of messages and posters on the [R-help] mailing list
• Estimating the number of R users around the world
• The number of R users on Facebook and Twitter

Besides this free chapter, Packt offered a 50% discount on the e-book format of this book for two weeks, that you can activate via the RXI37LH discount code until October 30 2015 (Friday). Another promo code for 20% discount on printed copies is also being generated — to be available early next week. For more details, revisit this page later and look for new comments, or follow me on Twitter:

After ~1001 sleepless nights, my #rstats book on #datascience is published w/ a free chapter https://t.co/7lS4pgN06k pic.twitter.com/MnM24P67dE

— Gergely Daróczi (@daroczig) October 1, 2015

Some quick statistics on the book:

• 14 chapters
• 396 pages
• 95 packages loaded
• hflights and data.table used in 7, ggplot2 in 5, dplyr and plyr in 4, microbenchmark and MASS used in 3 chapters
• 5 reviewers
• more than 20 persons contributing
• 2,711 lines of the code bundle on GitHub
• 581 days between signing the author contract and the actual publication date
• around 320 e-mail sent and received with the ISBN on the subject line
• 10,000 kilometers between the places where I wrote the first and the last chapters
• and I forgot to use time tracking software after logging 174.73 hours spent on the book

And most importantly: I’d love to and looking forward to hearing any kind of private or public feedback on this book!