Enable summary method for all currently implemented frequentist experiments

[drbenvincent]

• We can now call the summary method on all the existing frequentist experiments
• Notebooks have been updated to demonstrate the result.summary() functionality
• Closes SC + ITS: quantitative outputs #19
• Basic test coverage of the summary method just to get code coverage to pass. But note that Have more coherent testing of the summary method #305 exists to do better testing on this front

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 80.87%. Comparing base (4af4af6) to head (9fc0798).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #355      +/-   ##
==========================================
+ Coverage   79.96%   80.87%   +0.91%     
==========================================
  Files          21       21              
  Lines        1642     1668      +26     
==========================================
+ Hits         1313     1349      +36     
+ Misses        329      319      -10     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

juanitorduz commented on 2024-06-19T08:00:38Z
----------------------------------------------------------------

These numbers look in different columns. Why don't er print a dataframe?

[juanitorduz]

Some tables based on print statements look out of phase. What if we use something like in the pyfixest package https://py-econometrics.github.io/pyfixest/quickstart.html ?

[drbenvincent]

Some tables based on print statements look out of phase.

Very good point @juanitorduz . For the moment I've kept the same basic approach, but have improved the formatting. The function that prints the model coefficients now evaluates the length of the longest coefficient label so that it can align the coefficient values appropriately.

I'm very open to rethinking the nature of the summary outputs, but I'm tempted to deal with that in a separate issue. For example, there is already #174 as one option.

I'll do a similar formatting improvement for the pymc experiments in a different issue now.

What do you think?

PS. I just noticed I probably have to update the doctests 🤣 Will do that now :)

[juanitorduz]

ok! let's follow up the next iteration on the associated issue :)