The MODEL Procedure

The MODEL procedure now allows you to model a nonconstant error variance, permitting the specification of ARCH- and GARCH-type regression models.

[IMAGE] The following features have been added to the MODEL procedure in Version 8:

NOOLS and NO2SLS options specify bipassing using OLS or 2SLS to get initial parameter estimates for GMM, ITGMM, or FIML. This is important for certian models that are poorly defined in OLS or 2SLS or if good initial parameter values are already provided. Note that for GMM, the V matrix is created using the initial values specified and this may not be consistently estimated.
MSE. variables are now available for estimation and simulation. There is a MSE. variable created for each dependent/endogenous variable in the model. The MSE.y variable contains the value of the mean square error for y at each iteration. These variables are used to specify the lagged values for GARCH type models.
Quasi-random numbers can be used to drive Monte Carlo simulation. Quasi-random numbers are specified using the QUASI= option on the SOLVE command. Two Quasi-random number generators supported by the MODEL procedure are the Sobol sequence (QUASI=SOBOL) and the Faure sequence (QUASI=FAURE). The default is QUASI=NONE which is the psuedo random number generator.