christoffersen-lib 0.1.8

Christoffersen Independence Test for VaR backtesting

christoffersen-lib

Python library for the Christoffersen Independence Test, used in VaR (Value at Risk) backtesting in quantitative finance.

Installation

pip install christoffersen-lib

Usage

from christoffersen import independence_test

violations = [0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 1, 0]

lr, p_value = independence_test(violations)

print(f"LR statistic : {lr:.4f}")
print(f"P-value      : {p_value:.4f}")

if p_value > 0.05:
    print("✅ Independence not rejected (violations are not clustered)")
else:
    print("❌ Independence rejected (violations are clustered — bad VaR model)")

Interpretation

P-value	Conclusion
> 0.05	VaR model passes the independence test
≤ 0.05	Violations are clustered — model underestimates tail risk

Parameters

independence_test(violations)

• violations : list or numpy array of 0s and 1s — 1 means the VaR was exceeded on that day

Returns (lr_statistic, p_value).

Dependencies

• numpy
• scipy

License

MIT