Copula
Quantitative Finance glossary · Probability, Statistics & Linear Algebra
Definition
A multivariate CDF on [0,1]^d with uniform marginals that joins arbitrary marginals into a joint distribution: by Sklar's theorem F(x_1,ldots,x_d) = C(F_1(x_1),ldots,F_d(x_d)). Decouples marginal modelling from dependence modelling. The Gaussian copula was the workhorse of pre-2008 CDO pricing (Li 2000) and its failure to capture tail dependence — extreme defaults clustering far more than Gaussian implies — was a proximate cause of CDO mispricing in the GFC. Student-t and Clayton/Gumbel Archimedean copulas exhibit non-zero tail dependence.
Copula in plain English
Sometimes you understand each ingredient of a problem on its own — how one stock behaves, how another behaves — but the hard part is describing how they behave *together*, especially in a panic. A copula is a clever tool that lets you handle those two jobs separately: model each thing individually, then glue them together with a chosen 'togetherness' pattern. Its fame is a cautionary tale: before 2008, banks used one popular gluing pattern (the Gaussian copula) to price bundles of mortgages, but it badly underestimated how often many borrowers default all at once in a crisis. That blind spot — extreme events clustering far more than the model assumed — helped fuel the mispricing behind the financial crisis, and safer copulas (Student-t, Clayton, Gumbel) were designed to capture those simultaneous meltdowns.
Learn these first
Related Quantitative Finance terms
Quantitative Finance practice questions · Quantitative Finance flashcards · Quantitative Finance study guide