You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Stata is very concise and easy. But it's proprietary and old-fashioned. Let's future proof our work, make sure the kids can understand and play with it, including these young Artificial Intelligences.
This was mostly done in 69ae19f. This commit has all DataDownloads/ and Predictors/ translated and validated. We're just waiting on @idreesm18 to finish Placebos/.
The Signals/Code/Predictors/ branch was more carefully translated with the Stata -> Python conversion finalized here: 653c519. The predictors Stata comparison test results for this commit are here. This is the "final" predictors Stata comparison test in the sense that after this commit, we will start to address Github issues in the Python code only, and thus the Python code will start to deviate from the Stata code (though I kept updating this md file to track how things are changing).
The Signals/Code/Placebos branch is not finalized until after we started addressing Github issues, so the Stata comparison results we'll see there will not look best until after commit 653c519.
@idreesm18 finished up almost all of the Placebos. The remainder are finalized in 5637bd5. Test results can be found here.
Note that there are also test results for the predictors in this commit, but the Python predictors have been modified to address Github issues and no longer track Stata so closely.
Activity
chenandrewy commentedon Sep 5, 2025
This was mostly done in 69ae19f. This commit has all DataDownloads/ and Predictors/ translated and validated. We're just waiting on @idreesm18 to finish Placebos/.
chenandrewy commentedon Oct 7, 2025
The Signals/Code/Predictors/ branch was more carefully translated with the Stata -> Python conversion finalized here: 653c519. The predictors Stata comparison test results for this commit are here. This is the "final" predictors Stata comparison test in the sense that after this commit, we will start to address Github issues in the Python code only, and thus the Python code will start to deviate from the Stata code (though I kept updating this md file to track how things are changing).
The Signals/Code/Placebos branch is not finalized until after we started addressing Github issues, so the Stata comparison results we'll see there will not look best until after commit 653c519.
chenandrewy commentedon Oct 17, 2025
@idreesm18 finished up almost all of the Placebos. The remainder are finalized in 5637bd5. Test results can be found here.
Note that there are also test results for the predictors in this commit, but the Python predictors have been modified to address Github issues and no longer track Stata so closely.
I can't believe this is done 😮💨