[ENH] SubsetResampling issue #5. - #9

Reference Issues/PRs

Added SubsetResampling estimator for bagging on features of a general portfolio estimator as discussed in discussion #5

What does this implement/fix? Explain your changes.

A new estimator called SubsetResampling method as discussed in this blog post:

https://blog.thinknewfound.com/2018/07/machine-learning-subset-resampling-and-portfolio-optimization

Does your contribution introduce a new dependency? If yes, which one?

No.

What should a reviewer concentrate their feedback on?

Checking that the newly written estimator complies with all the properties of a proper portfolio estimator.

Did you add any tests for the change?

I added a test_subsetresampling under a new section _bagging in the proper test/test_optimization/test_ensemble/test_subsetresampling.py.
The test checks everything works with 10 independent features subsets using the minimum variance portfolio.

Any other comments?

Authors of the subset resampling method suggest to set max_features=n^0.8 where n in the total number of assets, this has not been added in the class documentation.

PR checklist

For all contributions

• I've added myself to the list of contributors
  How to: add yourself to the all-contributors file in the skfolio root directory (not the CONTRIBUTORS.md).
• Optionally, I've added myself and possibly others to the CODEOWNERS file - do this if you want to become the owner or maintainer of an estimator you added.

For new estimators

• I've added the estimator to the API reference in docs/api.rst: ✅
• I've added one or more illustrative usage examples to the docstring and the examples section. ❌