Metadata-Version: 2.4
Name: decision-security
Version: 0.1.0a2
Summary: Reusable decision-science utilities for security: Monte Carlo, Bayes, Survival, VoI, causal helpers, and viz.
Project-URL: Homepage, https://github.com/security-decision-science/decision-security
Project-URL: Repository, https://github.com/security-decision-science/decision-security
Project-URL: Issues, https://github.com/security-decision-science/decision-security/issues
Project-URL: Documentation, https://github.com/security-decision-science/security-decision-science-book
Author: Laura Voicu
License: MIT
License-File: LICENSE
Keywords: bayesian,causal,cybersecurity,monte carlo,risk,survival,value of information
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Security
Requires-Python: >=3.9
Requires-Dist: matplotlib>=3.6
Requires-Dist: numpy>=1.23
Requires-Dist: pandas>=1.5
Requires-Dist: scipy>=1.10
Provides-Extra: test
Requires-Dist: mypy>=1.8; extra == 'test'
Requires-Dist: pytest-cov>=4; extra == 'test'
Requires-Dist: pytest>=7; extra == 'test'
Requires-Dist: ruff>=0.4; extra == 'test'
Description-Content-Type: text/markdown

# Decision Security

Reusable **decision-science utilities for security** — Monte Carlo risk bands, Bayesian updates & calibration, survival helpers, Value of Information, light causal helpers, and visualization.

```bash
pip install decision-security 
```

## Quickstart

```python
import numpy as np
from decision_security.montecarlo import risk_bands, var_es, make_lognormal_severity, simulate_aggregate_losses

sev = make_lognormal_severity(meanlog=8.0, sdlog=1.2)
losses = simulate_aggregate_losses(n_periods=10000, lam=0.6, severity_sampler=sev)
print(risk_bands(losses))      # {'p50': ..., 'p90': ..., 'p95': ...}
print(var_es(losses))          # (VaR95, ES95)
```

## Modules
	•	synth: synthetic data (heavy-tail losses, counts, mixtures, survival with censoring, categorical/Dirichlet).
	•	montecarlo: Poisson frequency + severity, risk bands, VaR/ES.
	•	bayes: Beta-Binomial & Normal(known σ) updates, calibration helpers.
	•	survival: simple Kaplan–Meier & Nelson–Aalen estimates.
	•	voi: Expected Value of Perfect Information (EVPI) and simple ROI selection.
	•	causal: tiny DAG utilities (parents, descendants, naive backdoor set).
	•	viz: small matplotlib helpers (loss distribution, risk bands, KM curves).

Status: 0.x (APIs may change). MIT License.

See docs & examples: Security Decision Science Book and the Security Decision Labs playground.
