2. You could say I’ve been using
for a while…
–Johnny Appleseed
3. But then I decided to “grow up”
and learn something
“useful”
4. For Statisticians By Statisticians
This is a good thing…and a bad thing
R grew out of an appreciation for the S programming language
(fathered by John Chambers at Bell Labs)
Ross Ihaka and Robert Gentleman set out to make a better statistics
lab at U. of Aukland; R was the result
5. A Little Context
R, like S, is pretty Scheme-y
But it’s also OO, except when it’s not
And mostly people stitch packages together
until something works
6. The Packages You'll Need
For the budding data scientist,
bare minimum starter set…
• caret
• ggplot2 (ggvis)
• data.table
• dplyr
• RODBC
• lattice
• RCurl
• Hmisc/rms
• zoo
• reshape2
7. The Places You'll Go
The non-definitive list of resources
• http://www.r-bloggers.com/
• http://stackoverflow.com/questions/tagged/r
• http://www.statmethods.net/ (Quick R)
• http://onepager.togaware.com/ (One page R)
• http://www.rdocumentation.org/
8. The Hadleyverse
Hadley Wickam gets +3 charisma for just showing up…
• ggplot2
• lubridate
• reshape2
• stringr
• dplyr
• ggvis
9. Lasciate ogne speranza, voi ch'entrate
Oh, by the way…
• R has a perverse love for floating point numbers
• Repeat after me…map, don’t loop; map, don’t loop
• R is terse…use tiny functions and lots of them
• Expect inconsistency, skepticism is your friend