The following features have been added to the QLIM procedure:
Bayesian estimation features. Most of the multivariate models available in the QLIM procedure can now be estimated in a Bayesian framework by using the BAYES statement.
Endogeneity issues can now be addressed in most of the models available in the QLIM procedure:
Parameters of a single structural model, such as a discrete choice model or limited dependent variable model with endogenous explanatory variables, along with the parameters of the reduced form models can be estimated consistently. There is no restriction on the nature of the endogenous explanatory variables; they can be discrete, censored, or truncated.
A new endogeneity test for the explanatory variables is available.
An overidentifying restrictions type test is available to test the validity of the instruments.
For Heckman’s two-step procedure, a new option is available. This option enables the model for the selected sample to be a discrete choice or limited dependent variable model.