CQLIM Procedure

The CQLIM procedure analyzes univariate and multivariate limited dependent variable models in which dependent variables take discrete values or are observed within only a limited range of values. When the dependent variable takes discrete values, you can use either a logit model, a probit model, or an ordered logit or probit model, depending on the distribution of the dependent variable. In addition to the discrete variable models, the CQLIM procedure includes Tobit models, truncated regression models, and multivariate models. The multivariate model can contain discrete choice and limited dependent variables in addition to linear continuous dependent variables.

Another class of limited dependent variable models that the CQLIM procedure supports is stochastic frontier models. These models were first introduced by Aigner, Lovell, and Schmidt (1977) and Meeusen and van den Broeck (1977); they are capable of analyzing technical inefficiencies in production functions or cost functions.

The CQLIM procedure is similar to the QLIM procedure in SAS/ETS software; PROC CQLIM requires SAS Cloud Analytic Services (CAS) in order to run. Both procedures analyze univariate and multivariate limited dependent variable models. Unlike the QLIM procedure, which can run only on a single workstation, the CQLIM procedure takes advantage of the distributed computing environment of CAS, which enables it to distribute the optimization task to one or more nodes. In addition, each node can use one or more threads to perform the optimization on its subset of the data. When several nodes are used and each node uses several threads to carry out its part of the work, the result is a highly parallel computation that can provide a dramatic gain in performance.

The CQLIM procedure can use the maximum likelihood method or the Bayesian method to perform the inference of the available models. By default, it uses multiple threads to perform computations.