Stata integrations

[wbuchanan]

I know someone has already put together something for Stata integration with Jupyter and recently saw a similar project pop-up for integrating SAS with Jupyter notebooks. Just wondering if there is any other work happening with integrating other tools in the Stata environment.

[Matthew J. Baker]

wbuchanan -

While I don't know if there are any other integrations, I would like to giveipystata a ringing endorsement. It really works well for documenting projects, as the markup language is very flexible. The administrator of the project has been very helpful and has even added mata integration.

While it might seem strange to run Stata through Python in a workbook setting, there is a very nice benefit to doing this: the way ipystata works, one can easily open multiple sessions at the same time and move back and forth between Stata and Python. This is a really big advantage if you want to run things in parallel, or have a bunch of data sets open at once.

Best,

Matt

[wbuchanan]

Matthew J. Baker Jupyter is pretty phenomenal with regards to how it allows data to move across programming languages/environments without much effort. There's a somewhat related project that is tailored a bit more to Big Data stacks (e.g., Hadoop, Spark, etc...) built on the JVM called Zeppelin. It'd be interesting to see if anyone decides to try building out an IDE analogous to RStudio using the same development framework.

[Kyle Barron]

Just adding this to another Stata-Jupyter thread:

Along with Mauricio Caceres I wrote a Jupyter kernel for Stata that works on Windows, Linux, and Mac, and (unlike ipystata) controls Stata directly and thus doesn't need to slow down while moving data between Python and Stata.

You can get more information here: https://kylebarron.github.io/stata_kernel/