The MODEL Procedure

Overview

The MODEL procedure estimates and simulates systems of variables with correlated residuals. In prior releases, these residuals were restricted to have normal distributions with mean zero and non-constant variance. The following new distributions are now supported for multivariate estimation:

The centered t-distribution with estimated degrees of freedom and non-constant variance
A user-specified likelihood function

The following distributions are now supported for multivariate Monte Carlo simulation:

Cauchy
Chi-squared
Empirical
F
Poisson
t
Uniform

A transformation technique is used to create a covariance matrix for generating the correct innovations in a Monte Carlo simulation.

Chapter Contents
Previous
Next
Top