The MODEL statement specifies the dependent variable and independent regressor variables for the regression model.
You can specify the following option after a slash (/):
NOINT
suppresses the intercept parameter.
You can also specify the following endogenous variable options, which are the same as the options in the ENDOGENOUS statement. If you specify an endogenous variable option in both the MODEL statement and the ENDOGENOUS statement, only the option in the ENDOGENOUS statement is used.
Discrete Variable Options
DISCRETE <(discrete-options)>
specifies that the endogenous variables in the MODEL statement be discrete. The BAYES statement is not compatible with ordinal models—that is, discrete models that have more than two levels of the response variable. You can specify the following discrete-options:
DISTRIBUTION=LOGISTIC | NORMAL
DIST=LOGISTIC | NORMAL
D=LOGISTIC | NORMAL
specifies the cumulative distribution function to use to model the response probabilities. You can specify the following distribution types:
LOGISTIC
specifies the logistic distribution for the logit model.
NORMAL
specifies the normal distribution for the probit model.
By default, DISTRIBUTION=NORMAL.
If a multivariate model is specified, logistic distribution is not allowed. Only normal distribution is supported.
ORDER=FORMATTED | FREQ | INTERNAL
specifies the sort order for the levels of the discrete variables that you specify in the ENDOGENOUS statement. The sort order determines which parameters in the model correspond to each level in the data. You can specify the following sort orders:
FORMATTED
sorts levels by formatted value. The sort order is machine-dependent.
FREQ
sorts levels by descending frequency count; levels that have the most observations come first in the order.
INTERNAL
sorts levels by unformatted value. The sort order is machine-dependent.
By default, ORDER=FORMATTED. For more information about sort order, see the chapter on the SORT procedure in the Base SAS Procedures Guide.
Censored Variable Options
CENSORED <(censored-options)>
censors the endogenous variables in this statement. You can specify the following censored-options:
LB=value |variable
LOWERBOUND=value |variable
specifies the lower bound of the censored variables. If value is missing or the value of variable is missing, no lower bound is set. By default, no lower bound is set.
UB=value |variable
UPPERBOUND=value |variable
specifies the upper bound of the censored variables. If value is missing or the value of variable is missing, no upper bound is set. By default, no upper bound is set.
Truncated Variable Options
TRUNCATED <(truncated-options)>
truncates the endogenous variables in this statement. You can specify the following truncated-options:
LB=value |variable
LOWERBOUND=value |variable
specifies the lower bound of the truncated variables. If value is missing or the value of variable is missing, no lower bound is set. By default, no lower bound is set.
UB=value |variable
UPPERBOUND=value |variable
specifies the upper bound of the truncated variables. If value is missing or the value of variable is missing, no upper bound is set. By default, no upper bound is set.
Stochastic Frontier Variable Options
FRONTIER <(frontier-options)>
specifies the stochastic frontier model. You can specify the following frontier-options:
COST
specifies that the estimated model be a cost function.
PRODUCTION
specifies that the estimated model be a production function.
TYPE=EXPONENTIAL | HALF | TRUNCATED
specifies the model type.
EXPONENTIAL
specifies an exponential model.
HALF
specifies a half-normal model.
TRUNCATED
specifies a truncated normal model.
If neither PRODUCTION nor COST is specified, a production function is estimated by default. The BAYES statement is not compatible with stochastic frontier models.