Friedland Chapter 9 Bornhuetter-Ferguson Technique

[priyam0k]

Summary of Changes

Adds a first draft of Chapter 9 – Bornhuetter-Ferguson Technique, recreating the XYZ Insurer Auto BI exhibits from Friedland Chapter 9. Like Chapters 8 and 11, it's a notebook with reconciliation asserts built in.

Everything is computed in-package (Development + TailConstant + BornhuetterFerguson), reusing the Chapter 7 XYZ selection (volume-weighted 2-period; 1.000 reported / 1.010 paid tail; age-to-age factors rounded to 3 decimals per the text). It recreates both exhibits — Ultimate Claims Projection and Development of Unpaid Claim Estimate — and reconciles exactly: ultimate 513,145 / 554,543 and IBNR 63,519 / 104,917 (reported / paid).

misc:

• adds an Earned Premium column to the friedland_xyz_auto_bi sample (CSV + load_sample) so the a priori premium is read natively instead of hardcoded (same approach as Friedland Chapter 11 #1079).

Related GitHub Issue(s)

refers #588

Additional Context for Reviewers

• The Earned Premium change is additive; Chapters 6 and 7 only select Reported Claims/Paid Claims, and their doctests still pass (35/35 and 76/76).

• Chapter 9 is a notebook, so its asserts run at build (execute_notebooks: force) and a failure surfaces as a warning, per the Chapter 11 convention.

• Note: Friedland Chapter 11 #1079 is separately adding XYZ premium to the freq-sev sample — worth reconciling once it merges so premium isn't carried twice.

• I passed tests locally for code (pytest) and documentation (jb build docs --builder=custom --custom-builder=doctest)

---

Note

Low Risk
Documentation-only change (notebook + TOC); no application or library code in this diff.

Overview
Adds Friedland Chapter 9 to the docs as a new Jupyter notebook and links it in docs/_toc.yml under the Friedland section.

The notebook walks through the XYZ Insurer Auto BI Bornhuetter–Ferguson example using in-package Development, TailConstant, and BornhuetterFerguson (Chapter 7 development selection, a priori via sample_weight, reported CDF cap at 1.0). It rebuilds the ultimate projection and unpaid-claims exhibits and ends with numpy asserts that lock CDFs, ultimates, and IBNR to the printed Chapter 9 values.

^{Reviewed by Cursor Bugbot for commit 599e826. Bugbot is set up for automated code reviews on this repo. Configure here.}

[henrydingliu]

@priyam0k experimental is backed up. i created a clean experimental to prioritize your work. please resolve conflicts

[github-actions]

Pyright Type Completeness

View the full pyright --verifytypes output for this commit

Project (full chainladder package, at this PR's head): 14.5% of exported symbols fully typed (185 / 1272)

	Known	Ambiguous	Unknown	Total
Project (head)	185	106	981	1272

Other symbols referenced but not exported by chainladder: 13

	Known	Ambiguous	Unknown	Total
Other (head)	3	1	9	13

Symbols without documentation:

• Functions without docstring: 313
• Functions without default param: 0
• Classes without docstring: 10

Patch (exported symbols added or changed by this PR): no exported symbol type-completeness changes detected.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.50%. Comparing base (32ae098) to head (599e826).

Additional details and impacted files

@@              Coverage Diff              @@
##           experimental    #1109   +/-   ##
=============================================
  Coverage         90.50%   90.50%           
=============================================
  Files                91       91           
  Lines              5289     5289           
  Branches            671      671           
=============================================
  Hits               4787     4787           
  Misses              359      359           
  Partials            143      143           

Flag	Coverage Δ
unittests	90.50% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.