David X. Li (2000) Gaussian Copula CDO Tranche Pricing Lab

Price synthetic collateralized debt obligation tranches, evaluate equity upfronts and mezzanine/senior breakeven spreads, and explore default correlation smiles using David X. Li's (2000) seminal copula framework.

Structured Credit Presets:

Credit Pool & Copula Parameters

CDS Index Benchmark Spread (bps):

Average 5Y spread for 125 index obligors.

Asset Correlation ρ (%):

One-factor latent variable correlation.

Recovery Rate R (%):

Standard senior unsecured recovery.

Risk-Free Rate r (%):

Tranche Maturity T (Years):

Portfolio Notional ($):

Tranche Benchmark Pricing Summary

Hazard Rate (λ)

5Y Pool Loss

Equity [0% - 3%] Market Quote

Jr Mezz [3% - 7%]

Senior [10% - 15%]

Super Senior [15% - 30%] Spread

Structured Credit Visualizer

CDO Capital Structure & Tranche Valuation Waterfall

Tranche Segment	Notional ($)	Expected Loss (%)	Market Quote	Risky PV01	Protection Leg

Sensitivity: Correlation (ρ) vs Index CDS Spread on Equity Tranche Upfront (%)

ρ \ CDS	40 bps	65 bps	100 bps	160 bps	250 bps

Demonstrates that higher correlation reduces equity first-loss upfront fees.

Sensitivity: Correlation (ρ) vs Recovery Rate (R) on Jr Mezzanine [3% - 7%] Spread (bps)

ρ \ R	20%	30%	40%	50%	60%

Highlights how recovery rate and systemic correlation drive subordination breach risk.

Mathematical Foundations: David X. Li (2000) Gaussian Copula & CDO Tranching

1. Latent Asset Returns & Single-Factor Copula

David X. Li's (2000) breakthrough was mapping survival times $\tau_i$ to standard normal random variables $X_i = \Phi^{-1}(F_i(\tau_i))$. Under a one-factor Gaussian copula, each obligor's normalized return is driven by a common systematic market factor $M \sim \mathcal{N}(0, 1)$ and an idiosyncratic shock $Z_i \sim \mathcal{N}(0, 1)$:

$$X_i = \sqrt{\rho} M + \sqrt{1 - \rho} Z_i$$

Conditional on the systematic macroeconomic state $M = m$, default events are mutually independent Bernoulli trials with conditional default probability:

$$p(t \mid m) = \Phi\left( \frac{\Phi^{-1}(F(t)) - \sqrt{\rho} m}{\sqrt{1 - \rho}} \right)$$

In a Large Homogeneous Portfolio (LHP), the portfolio loss fraction conditional on $m$ is simply $L(t \mid m) = (1 - R) p(t \mid m)$.

2. Tranche Waterfall & Breakeven Spreads

A CDO tranche with attachment point $A$ and detachment point $D$ absorbs cumulative losses according to the piecewise linear tranche loss function:

$$L_{[A, D]}(t) = \frac{1}{D - A} \min\left(D - A, \max(0, L(t) - A)\right)$$

Equating the expected present value of the Protection Leg (default indemnification) with the Premium Leg (running spread coupons on remaining tranche principal) yields the breakeven par spread $s_{[A, D]}^*$:

$$s_{[A, D]}^* = \frac{\sum_k e^{-r t_k} \mathbb{E}[\Delta L_{[A, D]}(t_k)]}{\sum_k \Delta t_k e^{-r t_k} \left[ 1 - \frac{\mathbb{E}[L(t_k)] + \mathbb{E}[L(t_{k-1})]}{2} \right]}$$

Interactive CDO Tranche & Copula Mastery Quiz

Test your understanding of correlation sensitivity, Gaussian copula tail dependence limits, compound vs. base correlation, and equity tranche quoting conventions.

Frequently Asked Questions

David X. Li's (2000) model links the marginal default time distributions of individual debt obligors using a multivariate Gaussian copula. By projecting latent asset returns onto a single common market factor and obligor-specific shocks, it allows rapid analytical and numerical evaluation of correlated credit portfolio loss distributions and CDO tranche prices.

Higher default correlation decreases equity tranche risk (lowering required upfront fee) because obligors either all survive or all default together, giving equity a higher chance of full survival. Conversely, higher correlation increases senior and mezzanine tranche risk because it exposes them to catastrophic cluster defaults that breach higher subordination thresholds.

Compound correlation is the single uniform correlation parameter calibrated to match the market price of an individual tranche [A, D]. Because compound correlation varies across tranches (creating a 'correlation smile') and often yields multiple or non-existent mathematical solutions for mezzanine tranches, dealers developed Base Correlation, which prices tranches consistently as differences between standardized base tranches [0, K].

The Gaussian copula exhibits zero asymptotic tail dependence, meaning it mathematically assumes extreme co-defaults become infinitely improbable as default thresholds increase. During the 2007–2008 subprime crisis, home prices fell nationwide, causing systemic correlation to spike and triggering massive simultaneous co-defaults that wiped out AAA-rated senior tranches.

By standard market convention (e.g. CDX.NA.IG), equity tranches [0% - 3%] are quoted as an upfront payment percentage of notional plus a fixed running coupon of 500 basis points per annum. Mezzanine and senior tranches are quoted as running par spreads in basis points per annum with zero upfront fee.

Related Credit Risk & Quantitative Structured Finance Simulations

Vasicek (2002) Credit Portfolio Lab

ASRF credit loss distributions & Basel II/III IRB capital.

Jarrow-Lando-Turnbull (1997) Lab

Markov chain credit rating migration & spread curves.

CDS Pricing Calculator

Credit default swap spreads & hazard rate bootstrapping.

CreditMetrics (1997) Lab

Multi-obligor portfolio CreditVaR, rating migration & asset correlation.

Copula & Tail Dependence Lab

Clayton, Gumbel, Frank, and Student-t copulas.