Factor Analysis of Information Risk (FAIR) model written in Python.
This package endeavors to create a simple API for automating the creation of FAIR Monte Carlo risk simulations.
This is based in large part on:
- the Open FAIR™ Technical Standard published by the Open Group; and,
- Measuring and Managing Information Risk
"Open FAIR" is a trademark of the Open Group.
pyfair is available on PyPI. To use pyfair with your Python installation, you can run:
pip install pyfairDocumentation can be found at the Read the Docs site.
import pyfair
# Create using LEF (PERT), PL, (PERT), and SL (constant)
model1 = pyfair.FairModel(name="Regular Model 1", n_simulations=10_000)
model1.input_data('Loss Event Frequency', low=20, mode=100, high=900)
model1.input_data('Primary Loss', low=3_000_000, mode=3_500_000, high=5_000_000)
model1.input_data('Secondary Loss', constant=3_500_000)
model1.calculate_all()
# Create another model using LEF (Normal) and LM (PERT)
model2 = pyfair.FairModel(name="Regular Model 2", n_simulations=10_000)
model2.input_data('Loss Event Frequency', mean=.3, stdev=.1)
model2.input_data('Loss Magnitude', low=2_000_000_000, mode=3_000_000_000, high=5_000_000_000)
model2.calculate_all()
# Create metamodel by combining 1 and 2
mm = pyfair.FairMetaModel(name='My Meta Model!', models=[model1, model2])
mm.calculate_all()
# Create report comparing 2 vs metamodel.
fsr = pyfair.FairSimpleReport([model1, mm])
fsr.to_html('output.html')
{
"Loss Event Frequency": {
"low": 20,
"mode": 100,
"high": 900
},
"Loss Magnitude": {
"low": 3000000,
"mode": 3500000,
"high": 5000000
},
"name": "Regular Model 1",
"n_simulations": 10000,
"random_seed": 42,
"model_uuid": "b6c6c968-a03c-11e9-a5db-f26e0bbd6dbc",
"type": "FairModel",
"creation_date": "2019-07-06 17:23:43.647370"
}