Import the best IDA examples

[briatte]

… as 'bonus' demos?

https://github.com/briatte/ida

• Not all examples are very academic, some are just fun data
• It would be nice to turn the course into a website like I did back then, but it would be too much work to write up everything.
• Another solution would be to just update the code from IDA and republish that…

Candidates for Week 2 (functions)

• Writing up a BMI function
• Possible conflicts between roommates

Candidates for Week 3 (data)

• Reading Obama 2008 v 2012 from Google Sheets
• Factors, using WDI data
  • Not the best example, but having a bonus on factors would be helpful
  • Same goes for dates
• Reshaping and plotting data: Case-Shiller Index, and US homicides
  • Could also fit in Week 4, but might be better to illustrate pivots and aggregation
• UK polling accuracy, 1984-2016
  • Data stop in 2016 (and includes general election results!)
  • Still a good example of importing from Google Sheets + some wrangling

Candidates for Week 4 (plots)

• Ideological polarization in the US Congress
  • There's new data for that
• Plotting an ECDF -- Week 4?
• Building a US recessions plot, bit by bit -- Week 4?
  • Would require finding the original data again

Candidates for Week 8 (linear models)

• Predicting 100m sprint times
  • Data source is gone, but shouldn't be too hard to replace

Candidates for Week 12 (extra stuff)

Maps (covered by #34)

• Isarithmic maps!
  • The code never got posted, but there are instructions
  • Search GitHub for akima map examples?

Web scraping:

• Web scraping a Twitter account (Beijing air pollution), plus minimal time series stuff
• Web scraping FIDE chess rankings
• Web scraping + Herfindhal-Hirschman Index -- Week 12?
  • Data source still looks processable
  • Have it as a Web scraping + R Markdown example?

Text:

• Julian Assange speech(es?)

Candidates for unassigned weeks

Networks:

• My own stuff…

Time series:

• Smoothing Simpsons TV rankings + changepoints
• Detrending the Piketty/Saez top income data

Reproducible/open science:

• That rant deserved to exist in 2013, and still does in 2023…