Friedland Chapter 11

[henrydingliu]

Summary of Changes

Adding a first draft of Chapter 11, which showcases the recently added TriangleWeight and DisposalRate

Unlike Chapter 6 and 7, we are using jupyter notebook, which provides richer formatting. Testing is achieved through asserts built into the notebook, though failing any assert only results in a warning.

RTD version here.

Related GitHub Issue(s)

refers #588
closes #1073 #1024

Additional Context for Reviewers

other misc edits:

• adding precision to several gl insurer dataset (use severity x count to get more decimals on loss)

• adding incremental paid severity to xyz dataset

• removing friedland_xyz_disposal (redundant)

• nan_triangle no longer returns full triangle for all patterns

• w_ is now deleted inside _drop_subtriangles

• TriangleWeight now supports full triangles

• fit_OLS is fixed inside WeightedRegression; r-squared is added

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

---

Note

Medium Risk
Changes to nan_triangle, weighted OLS, and TriangleWeight on full triangles affect core triangle arithmetic and development weighting; large notebook adds maintenance surface but limited runtime risk.

Overview
Adds Friedland Chapter 11 documentation (notebook + TOC) that chains existing estimators—Development, Chainladder, TriangleWeight, DisposalRate, Trend, WeightedRegression—to reproduce frequency-severity exhibits, with notebook assert checks against reference figures.

Triangle / weights: nan_triangle no longer treats pattern triangles like full triangles (only is_ultimate gets the all-ones mask). _drop_subtriangles also clears w_. TriangleWeight can be fit on full triangles (full_triangle_, etc.) via an is_full branch in n_periods weighting; sparse inputs are forced to numpy for fitting.

Regression: WeightedRegression.fit_OLS uses weighted means for slope/intercept and exposes rsq_.

Sample data: Updates Friedland CSVs (reported claims, incremental paid severity, precision); merges disposal columns into friedland_xyz_auto_bi and removes friedland_xyz_disp; manifest entries adjusted accordingly.

Tests: TriangleWeight tests for drop-all-valuations error and full-triangle weight parity with partial triangles.

^{Reviewed by Cursor Bugbot for commit 901ded6. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

Pyright Type Completeness

View the full pyright --verifytypes output for this commit

Project (full chainladder package, at this PR's head): 15.0% of exported symbols fully typed (196 / 1308)

	Known	Ambiguous	Unknown	Total
Project (head)	196	111	1001	1308

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: 315
• Functions without default param: 0
• Classes without docstring: 10

Patch (exported symbols added or changed by this PR): 0.0% fully typed (0 / 3)

	Known	Ambiguous	Unknown	Total
Patch	0	1	2	3

Patch symbol details

Symbol	Status	Change
chainladder.utils.tests.test_utilities.test_triangleweight_drop_valuation_all	❌ unknown	new
chainladder.utils.tests.test_utilities.test_triangleweight_full_triangle	❌ unknown	new
chainladder.utils.weighted_regression.WeightedRegression.rsq_	⚠️ ambiguous	new

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 91.20%. Comparing base (8461cfb) to head (901ded6).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1176      +/-   ##
==========================================
+ Coverage   91.14%   91.20%   +0.06%     
==========================================
  Files          91       91              
  Lines        5365     5380      +15     
  Branches      681      683       +2     
==========================================
+ Hits         4890     4907      +17     
+ Misses        340      339       -1     
+ Partials      135      134       -1     

Flag	Coverage Δ
unittests	91.20% <100.00%> (+0.06%)	⬆️

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.

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit fe37feb. Configure here.}