Skip to content

Actuarial Use Cases

This library was built to tackle the three main pillars of actuarial pricing and reserving:

1. Declarative Pricing (Property & Casualty / General)

Use Case: You are building a pricing rater for Auto Insurance. You need a base rate, and you need to multiply that rate based on categorical factors (State) and continuous bands (Age). The Pain: Writing if state == "CA": premium *= 2.0 across 50 states and thousands of rows is slow and error-prone. The Solution: FactorModel 

from riskpy import FactorModel

model = FactorModel(initial_base_rate=1000.0)
# Categorical Strings
model.add_multiplier("vehicle_type", "Sedan", 1.0)
model.add_multiplier("vehicle_type", "Sports Car", 2.5)

# Continuous Numeric Bands
model.add_numeric_band_multiplier("driver_age", 16, 25, 2.0)
model.add_numeric_band_multiplier("driver_age", 26, 99, 1.0)

2. Stochastic Tail Risk (Reinsurance / Capital Modeling)

Use Case: You need to calculate the 99% Value-at-Risk (VaR) for a catastrophe book. You expect 5 hurricanes a year (Poisson) with an average cost of $10M each (Lognormal). The Pain: Python's random or even NumPy can struggle when managing massive nested severity/frequency loops efficiently inside a GUI. The Solution: MonteCarloSimulator 

from riskpy import MonteCarloSimulator

# Runs 1,000,000 simulations in C++ (Mersenne Twister; pass seed= for a reproducible run)
# Returns a Python list of the aggregated losses for plotting
results = MonteCarloSimulator(trials=1000000).simulate_aggregate_loss(
    expected_frequency=5.0,
    expected_severity_mu=15.0,  # log scale: median e^15 ≈ $3.3M, mean e^(μ+σ²/2) ≈ $10M
    severity_sigma=1.5
)

3. Core Math (Life & Annuity)

Use Case: You need to rapidly discount cash flows or look up mortality decrements to price a life annuity. The Pain: Re-writing robust financial formulas across different scripts. The Solution: ActuarialMath 

from riskpy import ActuarialMath

# $50k yearly payment for 20 years at 5% discount rate
pv = ActuarialMath.present_value(rate=0.05, periods=20, payment=50000)

# An illustrative banded mortality rate for an 85 year old — not a published table
q_x = ActuarialMath.lookup_mortality_rate(age=85)

For real mortality tables, annuities and reserves, use riskpy.life.