test: example from wiki

[paulz]

This pull request includes several changes to the examples/team_recommender/tests suite, focusing on statistical significance testing and sample size calculations. The most important changes include the addition of new functions, modifications to existing test functions, and the removal of a test file.

New functions and modifications:

• examples/team_recommender/tests/helpers.py: Added a new function is_statistically_significant to determine statistical significance.

Test function updates:

• examples/team_recommender/tests/test_helpers.py: Updated test_next_success_rate to use pytest.approx for better precision in assertions.
• examples/team_recommender/tests/test_helpers.py: Added parameterized tests for test_next_sample_size_with_1_failure and test_next_no_failure_sample_size_via_loop to cover more cases.
• examples/team_recommender/tests/test_helpers.py: Added a new test test_example_on_wiki to validate statistical significance and sample size calculations.

File removals and dependency updates:

• examples/team_recommender/tests/test_proportions_ztest.py: Removed the entire file as it is no longer needed.
• pyproject.toml: Removed the statsmodels dependency, which was used in the deleted test file.

[Copilot]

Pull Request Overview

A refactoring to correct statistical calculations in calculate_ztest and improve test clarity for sample size functions.

• Corrected the parameters passed to proportions_ztest in calculate_ztest.
• Renamed and introduced new sample size functions to handle both one failure and no failure cases with accompanying tests.
• Improved error messages and added a test case based on a wiki example.

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
examples/team_recommender/tests/test_helpers.py	Refactored sample size helper functions and updated tests with consistent naming.
examples/team_recommender/tests/test_proportions_ztest.py	Corrected calculate_ztest parameters and updated significance tests with a wiki example.

Comments suppressed due to low confidence (3)

examples/team_recommender/tests/test_proportions_ztest.py:126

• The error message does not match the test parameters; it refers to '0 out of 3' while the sample size is 1000. Please update the message to accurately reflect the test scenario.

assert is_statistically_significant(0.7, 0, 1000), "not significant result for 0 out of 3"

examples/team_recommender/tests/test_proportions_ztest.py:131

• The assertion expects a statistically significant result, but the error message suggests otherwise. Please align the error message with the expected outcome.

assert is_statistically_significant(0.7, 0, 10), "no improvement detected at 10"

examples/team_recommender/tests/test_proportions_ztest.py:139

• The error message implies a lack of improvement despite the assertion requiring statistical significance. Please update the message to match the intended check.

assert is_statistically_significant(0.97, 0, 100), "no improvement detected at 100"

[carl]

🐻 approved