[suryaramku]

Description

This PR fixes the IPW normalization in DoubleMLSSM by replacing sum(dtreat) and sum(dcontrol) with the total sample size when calculating the treatment and control normalization factors.

Previously, the normalization incorrectly scaled the IPW residual corrections by the treatment and control group proportions. Under correctly specified propensity and selection models, the expected value of each unnormalized IPW weight is 1, so the normalization should use the total sample size rather than the number of observations in each treatment group.

Reference to Issues or PRs

Fixes #426

PR Checklist

Please fill out this PR checklist (see our contributing guidelines for details).

• [ x] The title of the pull request summarizes the changes made.
• [ x] The PR contains a detailed description of all changes and additions.
• [ x] References to related issues or PRs are added.
• [] The code passes all (unit) tests.
• [] Enhancements or new feature are equipped with unit tests.
• [x ] The changes adhere to the PEP8 standards.

Testing notes:

• Ran python -m pytest doubleml/irm/tests/test_ssm.py -q
• Result: 64 tests passed, 0 failed.
• Full repository test suite has not been run.
• This PR is a bug fix, not a new feature; no new regression tests have been added.

[SvenKlaassen]

I will close this as this was already fixed in #427