GeneratedAssembled automatically from moduleDocs.ts. No human has reviewed this page.

risk_metrics

Concept Overview

Downside risk measures over a return series or a return panel: value at risk (the quantile at the given confidence level), expected shortfall (the mean loss beyond it), conditional drawdown at risk, and portfolio variance from a covariance matrix and a weight vector. Expected shortfall and CDaR are subadditive where VaR is not, which is why a risk budget built on VaR alone can be gamed by splitting one position across two sleeves.

When to Use

Use it for portfolio-level guardrails and risk budgets, and as the input when hcaa should allocate on tail risk rather than on variance. Prefer expected shortfall to VaR whenever the number will be summed across books. These are non-parametric estimates, so they need enough tail observations to mean anything: at 95% confidence a 200-observation sample rests on ten points. All of them are &self methods on a unit struct, and the _from_matrix variants take return panels.

Mathematical Foundations

VaR

$VaR_\alpha = -Q_\alpha(R)$

Expected Shortfall

$ES_\alpha = -E[R \mid R \le Q_\alpha(R)]$

Usage Examples

Rust

Compute VaR and ES

use openquant::risk_metrics::RiskMetrics;

let returns = vec![-0.02, 0.01, -0.005, 0.003, 0.004];

// These are &self methods on a unit struct, not associated functions: they need
// a receiver. `confidence_level` is the tail probability (0.05 = 95% VaR).
let metrics = RiskMetrics;
let var_95 = metrics.calculate_value_at_risk(&returns, 0.05)?;
let es_95 = metrics.calculate_expected_shortfall(&returns, 0.05)?;

println!("VaR(95%) = {var_95:.4}, ES(95%) = {es_95:.4}");

API Reference

Python API

risk.calculate_value_at_risk
risk.calculate_expected_shortfall
risk.calculate_conditional_drawdown_risk
risk.calculate_variance
risk.calculate_value_at_risk_from_matrix
risk.calculate_expected_shortfall_from_matrix
risk.calculate_conditional_drawdown_risk_from_matrix

Rust API

RiskMetrics::calculate_value_at_risk
RiskMetrics::calculate_expected_shortfall
RiskMetrics::calculate_conditional_drawdown_risk
RiskMetrics::calculate_variance

Risk Notes and Caveats

Non-parametric estimates need enough tail observations.
Use matrix variants for multi-asset return panels.