rj678 / pycausalmatch

Causal Impact of an intervention integrated with control group selection
MIT License
9 stars 3 forks source link

Refer to remapped df #3

Closed bartdegroot92 closed 3 years ago

bartdegroot92 commented 3 years ago

Hi there!

First of all, thanks a lot for setting up this library in Python. We're eager to start using it!

As I already saw you noted you're aware of the bugs and the potential issues with multiple markets, I still wanted to flag where it's breaking for us. I think you remap the columns in ip_df here, but don't reference the ip_df after. Therefore data is not yet in the expected format and you'd want to take ip_df. I'm not 100% sure though as the rest of the code is new to me, but I proposed changes for this issue so you could easily add it if you agree!

git example

Thanks again and know your work is being used :)

rj678 commented 3 years ago

Thank you for your interest in this library and especially for raising this PR @bartdegroot92 :) I will review this soon and get back.