The following statements are available in the NLIN procedure:
PROC NLIN <options>;
BOOTSTRAP </ options>;
BOUNDS inequality <, …, inequality>;
BY variables;
CONTROL variable <=values> <…variable <=values>>;
DER. parameter=expression;
DER. parameter.parameter=expression;
ID variables;
MODEL dependent = expression;
OUTPUT OUT=SAS-data-set keyword=names <…keyword=names>;
PARAMETERS
<parameter-specification> <, …, parameter-specification>
</ PDATA=SAS-data-set>;
PROFILE parameter <…parameter> </ options>;
RETAIN variable <=values> <…variable <=values>>;
The statements in the NLIN procedure, in addition to the PROC NLIN statement, are as follows:
requests bootstrap resampling and estimation
constrains the parameter estimates within specified bounds
specifies variables to define subgroups for the analysis
specifies the first or second partial derivatives
specifies additional variables to add to the output data set
defines the relationship between the dependent and independent variables (the mean function)
creates an output data set containing observation-wise statistics
identifies parameters to be estimated and their starting values
identifies parameters to be profiled
includes assignment statements, ARRAY statements, DO loops, and other program control statements. These are valid SAS expressions that can appear in the DATA step. PROC NLIN enables you to create new variables within the procedure and use them in the nonlinear analysis. These programming statements can appear anywhere in the PROC NLIN code, but new variables must be created before they appear in other statements. The NLIN procedure automatically creates several variables that are also available for use in the analysis. See the section Special Variables for more information.
The PROC NLIN , PARAMETERS , and MODEL statements are required.