I want to run python codes on the current variables of Stata using PyStata. For example:
- set obs 5
- gen a = _n
- python
- a
- end
But python does not recognize "a". Instead I should run:
- python
- a = [1, 2, 3, 4, 5]
- a
- end
How can I use the existed "a" instead of new coding?
Hello Seyyed Amir Yasin Ahmadi. I think you'll have better luck posting this kind of question to the Statalist forum. Please take a look at the "advice on posting to Statalist" (second link below) first though. ;-)
- https://www.statalist.org/forums/forum/general-stata-discussion/general
- https://www.statalist.org/forums/help
Bruce Weaver
Thank you for your suggestion
https://www.statalist.org/forums/forum/general-stata-discussion/general/1737041-how-to-use-stata-generated-variables-in-pystata
Thanks for the link, Seyyed Amir Yasin Ahmadi. I have subscribed to that Statalist thread in hopes of learning something myself. ;-)
To use Stata variables in Python, you can leverage the pandas library, a powerful tool for data manipulation and analysis. Begin by installing the necessary packages, including pandas, and then read your Stata file into a DataFrame using pd.read_stata(). Accessing Stata variables is achieved by referring to them as columns in the DataFrame, allowing you to perform various data analyses and manipulations using pandas functions. After making modifications, if needed, save the updated data back to a Stata file with the to_stata method. While there isn't a specific package called PyStata, this approach enables seamless interaction with Stata variables in a Python environment.
Hello Alhaji Modu Isa. First, let me be clear that I am not a Pythonista, so I am not in a position to comment (very much) on your helpful looking reply. But I do wonder what you meant when you said "there isn't a specific package called PyStata". I wonder if you meant that the solution you offered does not use the PyStata package, rather than there is no such thing as a PyStata package. The reason I wonder that is because according to Stata's website, there is a "pystata" package. Here are some relevant links:
- https://www.stata.com/python/pystata18/
- https://www.stata.com/python/pystata18/api.html#ref-pystata
- https://www.stata.com/python/pystata18/notebook/Example0.html
From that last examples page:
This section provides a series of examples demonstrating how to call Stata from within Python by using the magic commands and API functions defined in the pystata Python package. For more details on their usage, you can go to Magic commands and API functions.
As I said, I am not a Pythonista. But that sounds to me as if it is relevant to anyone wanting to call Stata from within Python, and to use Stata variables when doing so.
Finally, the documentation I linked to is for Stata 18. I don't know if the pystata Python package was available prior to that version.
Thanks Bruce Weaver
According to your comment, I found that PyStata was for calling Stata from within Python, but I thought vise versa. I meant running Python in Stata. I think I have wrongly used PyStata term.
Thanks Alhaji Modu Isa
Pandas is a good idea. But I think there is a limitation that when we update the current Stata dataset we should run read_stata() again. I was looking for way for direct use of Stata variables.
However, it seems that it is an interesting option in Stata for simultaneous use of Stata and Python.
I received a ling from @Tim Huegerich in Stata blog
https://blog.stata.com/2020/11/05/stata-python-integration-part-8-using-the-stata-function-interface-to-copy-data-from-stata-to-python/
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression