Understanding the state of R in South Africa
summary
Downloads from CRAN are parsed from 1,402 log files (6 GB compressed) into a time series of 5.7 million observations (2.8M in 2018), decomposed with the stats package using the HTL function to show a strong upward trend from 2016 and a large 2018 spike; weekday patterns peak Mondays and slump weekends. Package analysis finds 14,156 unique packages with mean 203 and median 3 downloads, and an undirected package co‑download network pruned at edges >=100 reveals central hubs; an interactive visualization and a Shiny-based classify workflow are produced.
AI-written summary · about
chapters (9)
- 0:00Introduction and background
- 1:09Motivation for R usage analysis
- 3:36Data collection and processing
- 6:26Time series decomposition results
- 8:58Trends in R adoption drivers
- 9:25Package download statistics
- 11:16Network analysis approach
- 13:03Network visualization insights
- 16:26Key takeaways and future outlook
related talks
20m15 Years of R in Quantitative Finance
20mThe next million R users
40mPHP and R - Putting the PoweR in PHP Statistics
30m10 Billion Downloads: Insights and Trends in Open Source
23mGrowth Hacking - Product Analytics at Scale using R and RStudio
42mGetting started with R when analysing Github events
21mA guide to modern reproducible data science with R
15mStatically Analyzing the Dataflow of R Programs