[Data] Add actsim_sample: synthetic multi-D monthly triangle (#1218)

[jzhng105]

Adds actsim_sample, a synthetic incremental OMDM triangle with a two-level index, for testing and demonstrating multi-D / multi-grain workflows. All other grains are available via .grain() (e.g. OQDQ, OYDY) and 2-D/3-D views via .sum() / .groupby().

Shape: (9, 2, 120, 120)
Index: Region x LOB
Region: New_York, Chicago, Atlanta
LOB: COLL (collision), TPL_BI (third-party BI), AB (accident benefits)
Columns: reportedCount, Incurred (incremental)
Origin: monthly, 2016-01 to 2025-12; valuation 2025-12-31

Generated with ActSim (ClaimSimulator), seeds 1000-1008 (one per segment).

Exposure / frequency

• One annual-term policy cohort written each month, 2015-01 to 2025-12, 100 policies per cohort per region.
• Claim counts per cohort ~ Poisson(100 x freq): COLL 0.11 (lambda = 11), TPL_BI 0.010 (lambda = 1.0), AB 0.007 (lambda = 0.7)
• Accident dates uniform within the policy term (NHPP with alpha = 0); claims from accidents before 2016-01-01 are dropped.

Severity (lognormal, mu/sigma of log-loss, at 2016 level)

• COLL (8.3, 0.75): mean ~5.3k
• TPL_BI (9.9, 1.15): mean ~38.6k
• AB (10.2, 1.45): mean ~77.0k
• Trend: mu + 0.05 per accident year (~5.1% p.a. severity inflation).

Development (per-claim age-to-age factors keyed by development month, each perturbed by N(1, 0.10))

• COLL: {0: 1.6, 1: 1.3, 2: 1.05, 3: 1.0} ultimate by month 3
• TPL_BI: {3: 2.0, 12: 1.5, 24: 1.15, 36: 1.05, 48: 1.0} ultimate by month 48
• AB: {3: 2.3, 12: 1.8, 24: 1.4, 48: 1.15, 72: 1.05, 96: 1.0} ultimate by month 96
• reportedCount is recorded at each claim's first development age.

Known characteristics

• Incurred only moves at the keyed ages, so the TPL_BI and AB rows change in steps rather than every month.
• Multiplicative LDF noise produces some small negative incremental incurred amounts, mostly at the final development age.
• All figures are fictional.

Summary of Changes

Added actsim_sample, a synthetic incremental OMDM triangle with a two-level index, for testing and demonstrating multi-D / multi-grain workflows. All other grains are available via .grain() (e.g. OQDQ, OYDY) and 2-D/3-D views via .sum() / .groupby().

Related GitHub Issue(s)

ISSUE #1218 closed

AI/LLM Usage

Additional Context for Reviewers

Submitter's Checklist

• I have reviewed and am adhering to the standards outlined in the project Governing Doc.
• The PR subject title summarizes the changes, with one proper prefix ([FIX], [FEAT], [DOCS], [TST], [CHORE], or [BRK]).
• I am a human (not a bot), and this PR form is written by a human.

Reviewer's Checklist

• The implementation addresses the associated issue(s).
• The implementation is appropriate, maintainable, and follows ARCHITECTURE.md.
• PR subject title has the proper prefix and the subject is appropriate.
• Relevant issue(s) are linked.
• AI/LLM usage is disclosed and appropriate.
• Documentation and tests are appropriate.
• CI tests passed, or any failures are acceptable.
• Leave a comment with the final recommendation (e.g. approve as is, request a secondary review, or flag an area for more review).

[github-actions]

Pyright Type Completeness

View the full pyright --verifytypes output for this commit

Project (full chainladder package, at this PR's head): 15.6% of exported symbols fully typed (221 / 1420)

	Known	Ambiguous	Unknown	Total
Project (head)	221	108	1091	1420

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: 326
• Functions without default param: 0
• Classes without docstring: 8

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

[kennethshsu]

@jzhng105 please complete the PR template and resolve CICD checks

[henrydingliu]

instructions for adding new datasets is here.

https://github.com/casact/chainladder-python/blob/main/chainladder/utils/data/_manifest.py

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 92.59%. Comparing base (e4e0695) to head (8390fe3).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1425   +/-   ##
=======================================
  Coverage   92.59%   92.59%           
=======================================
  Files         102      102           
  Lines        5846     5846           
  Branches      723      723           
=======================================
  Hits         5413     5413           
  Misses        312      312           
  Partials      121      121           

Flag	Coverage Δ
unittests	92.59% <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.

[jzhng105]

Checked off the lists and registered the sample data in manifest

[kennethshsu]

@jzhng105 sorry, are you sure you read the checklist? PR template prefix in this case should be [FEAT], can you update that please?

For the future, please also fill out the AI/LLM Usage portion. I think we are ok for this one as I understand the data is completely synthetic.

[kennethshsu]

can we just use actsim here and drop the sample so everything looks a bit more uniformed?

[kennethshsu]

Please update the csv filename too! Thanks @jzhng105!!