ACTUNEO: African Actuarial Python Library

Vision and Objective

ACTUNEO is an open-source, community-driven actuarial Python library that empowers African and Zimbabwean actuaries to perform core actuarial, financial, and statistical computations with ease. The goal is to build a localized yet globally compatible toolkit that supports insurance, pensions, and investment analytics, while integrating with modern data science tools.

Status

ACTUNEO is an early-stage library. It has not been reviewed by an independent actuary and has not yet been run on a real insurer's data. Some parts reproduce published figures; others are checked only by hand calculation. It is not certified for statutory or audited reporting: check its output independently before relying on it.

Read Status and Limitations before using it. It lists what is implemented, how each part was checked, and what is missing.

Features

Implemented Modules

• mortality: Mortality tables, select mortality, life contingencies and commutation functions, with the Zimbabwe 2023 tables and AM92 included
• finance: Interest theory, yield curve construction/interpolation, duration, and convexity
• life: Life assurance, annuities, net and gross premiums and reserves with expenses, with-profits benefits, mortality profit, multiple decrement and multi-state models, unit-linked policies and profit testing
• loss_reserving: Claims triangles, chain-ladder (basic and inflation-adjusted), average cost per claim, Bornhuetter-Ferguson, Mack's standard error and bootstrap
• ifrs17: Premium allocation approach, liability for incurred claims with discounting and risk adjustment, reinsurance held, and the IFRS 17 statement of profit or loss, financial position and cash flows
• simulation: Aggregate claims simulation, and a claims simulator that produces triangles with a known true cost for testing reserving methods
• macro_africa: World Bank economic indicators (inflation, growth, exchange and interest rates) for African countries, downloaded on request
• zimbabwe: Catalogue of the Zimbabwe 2023 mortality tables and ZWL to ZiG conversion

In Development (Scaffolding Present)

• pensions: Contribution schedules, benefit projections, and actuarial valuations for pension schemes
• ifrs17 (further): General measurement model (GMM) and variable fee approach (VFA), with the contractual service margin
• utils: Data input-output helpers, validation, and reporting utilities

African Market Focus

• Localized Assumptions (Roadmap): Support for market-specific mortality, inflation, and interest rate tables
• Regulatory Alignment (Roadmap): Extensible templates for regulatory reporting (e.g., IPEC Zimbabwe, PASA, SAM)
• Currency Handling (Roadmap): Multi-currency modelling (USD, ZWL, Rand, etc.) with inflation-adjusted projections
• Socioeconomic Context (Roadmap): Assumptions relevant to informal sector, microinsurance, and low-coverage environments

Installation

From PyPI

pip install --pre actuneo

The current release is a beta (0.4.0b1), so --pre is needed: without it pip installs the old 0.1.2 release, whose calculations have since been corrected. Read the Status and Limitations page before relying on any result.

From Source

git clone https://github.com/ShannonT20/ACTUNEO.git
cd ACTUNEO
pip install -e .

Development Installation

pip install -e ".[dev]"

Quick Start

Mortality Analysis

from actuneo.mortality import MortalityTable, SurvivalFunctions

# Zimbabwe 2023 mortality tables are included
print(MortalityTable.zimbabwe_2023_tables())
mt = MortalityTable.from_zimbabwe_2023("male_assured_lives")

le = mt.life_expectancy(30)
print(f"Life expectancy at age 30: {le:.1f} years")

# Survival and life contingencies at 8% interest
sf = SurvivalFunctions(mt, interest_rate=0.08)
print(f"20-year survival probability from 30: {sf.npx(30, 20):.3f}")
print(f"20-year term assurance, A30:20 = {sf.assurance(30, 20):.5f}")
print(f"20-year annuity-due, a30:20 = {sf.annuity_due(30, 20):.4f}")

Financial Calculations

from actuneo.finance import InterestTheory, YieldCurve

# Interest theory
it = InterestTheory(interest_rate=0.05)
fv = it.future_value(1000, 10)  # Future value of $1000 in 10 years
print(f"Future value: ${fv:.2f}")

# Yield curve
maturities = [1, 2, 5, 10, 30]
yields = [0.03, 0.035, 0.045, 0.055, 0.065]
yc = YieldCurve(maturities, yields)
yield_15y = yc.get_yield(15)
print(f"15-year yield: {yield_15y:.3%}")

Life Insurance Calculations

Values are per unit sum assured; premiums are annual in advance.

from actuneo.life import LifeAssurance

la = LifeAssurance(mt, interest_rate=0.08)
sum_assured = 100_000

premium = sum_assured * la.net_annual_premium(30, 20, "endowment")
print(f"20-year endowment, annual premium at age 30: ${premium:,.2f}")

reserve = sum_assured * la.reserve_endowment(30, 20, 5)
print(f"Reserve after 5 years: ${reserve:,.2f}")

Claims Triangles and Chain-Ladder

from actuneo.loss_reserving import MackChainLadder, load_raa

triangle = load_raa()                  # or Triangle(...) / Triangle.from_long(...)
print(triangle.age_to_age().round(3))  # link ratios and their averages

mack = MackChainLadder(triangle, est_sigma="mack")
print(mack.summary().round(3))
print(f"IBNR {mack.total_ibnr:,.0f} with standard error {mack.total_mack_se:,.0f}")

Documentation

📚 Full documentation is available at https://actuneo.readthedocs.io/

The documentation includes:

• Installation Guide: Step-by-step installation instructions
• Quick Start: Get started with ACTUNEO in minutes
• Examples: Worked examples for the implemented modules
• Status and Limitations: What is and is not implemented, and how it was checked
• API Reference: Complete API documentation for all classes and functions
• Contributing Guide: How to contribute to the project
• Changelog: Version history and release notes

Documentation Sections

• Installation Guide
• Quick Start Tutorial
• Examples
• API Reference
• Contributing

Contributing

We welcome contributions from the actuarial community, especially from African and Zimbabwean actuaries and students. Here's how you can contribute:

Development Setup

1. Fork the repository
2. Clone your fork: git clone https://github.com/ShannonT20/ACTUNEO.git
3. Create a virtual environment: python -m venv venv
4. Activate the environment: venv\Scripts\activate (Windows) or source venv/bin/activate (Unix)
5. Install development dependencies: pip install -e ".[dev]"
6. Run tests: pytest

Guidelines

• Follow PEP 8 style guidelines
• Write comprehensive tests for new features
• Update documentation for API changes
• Add examples for new functionality
• Ensure cross-platform compatibility

Areas for Contribution

• African Mortality Tables: Contribute localized mortality data and improvement models
• Regulatory Frameworks: Implement calculations for specific African regulatory requirements
• Pension Systems: Develop models for various pension scheme structures
• IFRS 17: Implement insurance contract measurement models
• Loss Reserving: Add stochastic reserving methods
• Documentation: Improve documentation and add tutorials
• Testing: Expand test coverage and add integration tests

Roadmap

Phase 1 (Current): Core Implementation

• Basic package structure
• Mortality tables and survival functions
• Financial calculations (interest, yield curves)
• Life assurance and annuity calculations
• Unit testing framework
• Documentation setup

Phase 2: Expansion

• Pension calculations
• IFRS 17 implementation
• Loss reserving methods
• African economic data connectors
• Stochastic simulation tools

Phase 3: Advanced Features

• Machine learning integration
• Web application interfaces
• Regulatory reporting tools
• Multi-language support

Testing

Run the test suite:

pytest

Run with coverage:

pytest --cov=actuneo --cov-report=html

License

This project is licensed under the MIT License - see the LICENSE file for details.

Citation

If you use ACTUNEO in your research or work, please cite:

@software{sikadi_actuneo_2025,
  author = {Sikadi, Shannon Tafadzwa},
  title = {ACTUNEO: African Actuarial Python Library},
  url = {https://github.com/ShannonT20/ACTUNEO},
  year = {2025}
}

Contact

• Author: Shannon Tafadzwa Sikadi
• Email: shannonsikadi@gmail.com
• GitHub: https://github.com/ShannonT20/ACTUNEO
• LinkedIn: https://www.linkedin.com/in/shannon-sikadi-9370b3196/

Acknowledgments

• Inspired by existing actuarial libraries like lifecontingencies, actuarialmath, and pandas
• Special thanks to the African actuarial community for their support and feedback
• Built with modern Python data science tools (NumPy, Pandas, SciPy, Matplotlib)

ACTUNEO: Empowering African actuaries through open-source technology.