Added baseball.csv

[chrisjordansquire]

This seems to work. Was that what you had in mind?

[wesm]

That's cool. I was thinking about including a bigger piece to write meaningful/interesting groupby examples. I haven't looked into the license for the baseball data, plyr is MIT license so I assume that data is just public domain but...

[chrisjordansquire]

Didn't think about that.

It comes from http://www.baseball-databank.org/

They say their data is open for use by individual researchers, but other uses require their permission. On the other hand, I read the MIT license as saying copying any part of the software and its documentation is fine, and the data was released with the software. So it seems like the only real question is if the MIT license should be included with the baseball.csv file.

(The MIT license only says it needs repeating when 'substantial' parts of the software are copied.)

[chrisjordansquire]

Another potential solution is that I'm currently writing a little function to generate arbitrarily sized biostatistics-like data sets. I'd been intending it as a learning tool to let me explore pandas functionality for datasets more like what I'm used to (e.g., lots of categorical data), but it could be adapted to creating large tables of data for this as well.

[wesm]

I included a little chunk (100 rows) in doc/data so we should be all set