#
The COPULA Procedure (Experimental)

The COPULA procedure uses the copula concept to fit and simulate from multivariate distributions.
The copula concept enables you to express joint multivariate distributions in terms of their marginal
distributions and the dependency structure (correlations). Copula methods are popular in many different
areas (including risk management, credit scoring, asset allocations, and actuarial sciences) where many
different correlated factors must be modeled jointly. Copulas often help to perform large-scale multivariate
simulations of random vectors that would be difficult to perform using other multivariate fitting and
simulation methods.

The following are highlights of the COPULA procedure's capabilities:

- estimate the parameters for a specified copula type
- simulate a given copula
- plot dependence relationships among the variables
- supports the following types of copulas:
- normal copula
- Student's
*t* copula
- Clayton copula
- Gumbel copula
- Frank copula

- obtains separate analyses on observations in groups
- provides options to control the nonlinear optimization
- enables you to output data and estimates that can be used in other analyses
- supports ODS Graphics

## Documentation

For further details, see the *SAS/ETS*^{®} User's Guide

## Examples