The MODEL Procedure
Functional Summary
The statements and options in the MODEL procedure are summarized in Table 1.
Table 1: PROC MODEL Functional Summary
| Description | Statement | Option |
|---|---|---|
| Data Set Options | ||
| Specifies the input data set for the variables | FIT, SOLVE | DATA= |
| Specifies the input data set for parameters | FIT, SOLVE | ESTDATA= |
| Specifies the method for handling missing values | FIT | MISSING= |
| Specifies the input data set for parameters | MODEL | PARMSDATA= |
| Requests that the procedure produce graphics via the Output Delivery System | MODEL | PLOTS= |
| Specifies the output data set for residual, predicted, or actual values | FIT | OUT= |
| Specifies the output data set for solution mode results | SOLVE | OUT= |
| Writes the actual values to OUT= data set | FIT | OUTACTUAL |
| Selects all output options | FIT | OUTALL |
| Writes the covariance matrix of the estimates | FIT | OUTCOV |
| Writes the parameter estimates to a data set | FIT | OUTEST= |
| Writes the parameter estimates to a data set | MODEL | OUTPARMS= |
| Writes the observations used to start the lags | SOLVE | OUTLAGS |
| Writes the predicted values to the OUT= data set | FIT | OUTPREDICT |
| Writes the residual values to the OUT= data set | FIT | OUTRESID |
| Writes the covariance matrix of the equation errors to a data set | FIT | OUTS= |
| Writes the S matrix used in the objective function definition to a data set | FIT | OUTSUSED= |
| Writes the estimate of the variance matrix of the moment generating function | FIT | OUTV= |
| Reads the covariance matrix of the equation errors | FIT, SOLVE | SDATA= |
| Reads the covariance matrix for GMM and ITGMM | FIT | VDATA= |
| Specifies the name of the time variable | FIT, SOLVE, MODEL | TIME= |
| Selects the estimation type to read | FIT, SOLVE | TYPE= |
| General ESTIMATE Statement Options | ||
| Specifies the name of the data set in which the estimate of the functions of the parameters are to be written | ESTIMATE | OUTEST= |
| Writes the covariance matrix of the functions of the parameters to the OUTEST= data set | ESTIMATE | OUTCOV |
| Prints the covariance matrix of the functions of the parameters | ESTIMATE | COVB |
| Prints the correlation matrix of the functions of the parameters | ESTIMATE | CORRB |
| Printing Options for FIT Tasks | ||
| Prints the modified Breusch-Pagan test for heteroscedasticity | FIT | BREUSCH |
| Prints the Chow test for structural breaks | FIT | CHOW= |
| Prints collinearity diagnostics | FIT | COLLIN |
| Prints the correlation matrices | FIT | CORR |
| Prints the correlation matrix of the parameters | FIT | CORRB |
| Prints the correlation matrix of the residuals | FIT | CORRS |
| Prints the covariance matrices | FIT | COV |
| Prints the covariance matrix of the parameters | FIT | COVB |
| Prints the covariance matrix of the residuals | FIT | COVS |
| Prints Durbin-Watson d statistics | FIT | DW |
Prints first-stage  statistics | FIT | FSRSQ |
| Prints Godfrey’s tests for autocorrelated residuals for each equation | FIT | GODFREY |
| Prints Hausman’s specification test | FIT | HAUSMAN |
| Prints tests of normality of the model residuals | FIT | NORMAL |
| Prints the predictive Chow test for structural breaks | FIT | PCHOW= |
| Specifies all the printing options | FIT | PRINTALL |
| Prints White’s test for heteroscedasticity | FIT | WHITE |
| Options to Control FIT Iteration Output | ||
| Prints the inverse of the crossproducts Jacobian matrix | FIT | I |
| Prints a summary iteration listing | FIT | ITPRINT |
| Prints a detailed iteration listing | FIT | ITDETAILS |
| Prints the crossproduct Jacobian matrix | FIT | XPX |
| Specifies all the iteration printing-control options | FIT | ITALL |
| Options to Control the Minimization Process | ||
| Specifies the convergence criteria | FIT | CONVERGE= |
| Selects the Hessian approximation used for FIML | FIT | HESSIAN= |
| Specifies the local truncation error bound for the integration | FIT, SOLVE, MODEL | LTEBOUND= |
| Specifies the maximum number of iterations allowed | FIT | MAXITER= |
| Specifies the maximum number of subiterations allowed | FIT | MAXSUBITER= |
| Selects the iterative minimization method to use | FIT | METHOD= |
| Specifies the smallest allowed time step to be used in the integration | FIT, SOLVE, MODEL | MINTIMESTEP= |
| Modify the iterations for estimation methods that iterate the S matrix or the V matrix | FIT | NESTIT |
| Specifies the smallest pivot value | MODEL, FIT, SOLVE | SINGULAR |
| Specifies the number of minimization iterations to perform at each grid point | FIT | STARTITER= |
| Specifies a weight variable | WEIGHT | |
| Options to Read and Write Model Files | ||
| Deletes a model from a model file | DELETEMODEL | MODNAME= |
| Reads a model from one or more input model files | INCLUDE | MODEL= |
| Suppresses the default output of the model file | MODEL, RESET | NOSTORE |
| Specifies the name of an output model file | MODEL, RESET | OUTMODEL= |
| Deletes the current model | RESET | PURGE |
| Options to List or Analyze the Structure of the Model | ||
| Identifies equations in a dependency analysis | EQGROUP | |
| Identifies variables in a dependency analysis | VARGROUP | |
| Prints a dependency analysis of a simulation model | SOLVE | ANALYZEDEP= |
| Prints a dependency structure of a normal form model | MODEL | BLOCK |
| Prints a graph of the dependency structure of a normal form model | MODEL | GRAPH |
| Prints the model program and variable lists | MODEL | LIST |
| Prints the derivative tables and compiled model program code | MODEL | LISTCODE |
| Prints a dependency list | MODEL | LISTDEP |
| Prints a table of derivatives | MODEL | LISTDER |
| Prints a cross-reference of the variables | MODEL | XREF |
| General Printing Control Options | ||
| Expands parts of the printed output | FIT, SOLVE | DETAILS |
| Prints a message for each statement as it is executed | FIT, SOLVE | FLOW |
| Selects the maximum number of execution errors that can be printed | FIT, SOLVE | MAXERRORS= |
| Requests a comprehensive memory usage summary | FIT, SOLVE, MODEL, RESET | MEMORYUSE |
| Selects the number of decimal places shown in the printed output | FIT, SOLVE | NDEC= |
| Suppresses the normal printed output | FIT, SOLVE | NOPRINT |
| Turns off the NOPRINT option | RESET | |
| Specifies all the noniteration printing options | FIT, SOLVE | PRINTALL |
| Prints tables which summarize missing value calculations | FIT, SOLVE, MODEL | REPORTMISSINGS |
| Prints the result of each operation as it is executed | FIT, SOLVE | TRACE |
| Statements That Declare Variables | ||
| Associates a name with a list of variables and constants | ARRAY | |
| Declares a variable to have a fixed value | CONTROL | |
| Declares a variable to be a dependent or endogenous variable | ENDOGENOUS | |
| Declares a variable to be an independent or exogenous variable | EXOGENOUS | |
| Specifies identifying variables | ID | |
| Assigns a label to a variable | LABEL | |
| Selects additional variables to be output | OUTVARS | |
| Declares a variable to be a parameter | PARAMETERS | |
| Forces a variable to hold its value from a previous observation | RETAIN | |
| Declares a model variable | VAR | |
| Declares an instrumental variable | INSTRUMENTS | |
| Omits the default intercept term in the instruments list | INSTRUMENTS | NOINT |
| General FIT Statement Options | ||
| Omits parameters from the estimation | FIT | DROP= |
| Associates a variable with an initial value as a parameter or a constant | FIT | INITIAL= |
| Bypasses OLS to get initial parameter estimates for GMM, ITGMM, or FIML | FIT | NOOLS |
| Bypasses 2SLS to get initial parameter estimates for GMM, ITGMM, or FIML | FIT | NO2SLS |
| Specifies the parameters to estimate | FIT | PARMS= |
| Requests confidence intervals on estimated parameters | FIT | PRL= |
| Selects a grid search | FIT | START= |
| Options to Control the Estimation Method Used | ||
| Specifies nonlinear ordinary least squares | FIT | OLS |
| Specifies iterated nonlinear ordinary least squares | FIT | ITOLS |
| Specifies seemingly unrelated regression | FIT | SUR |
| Specifies iterated seemingly unrelated regression | FIT | ITSUR |
| Specifies two-stage least squares | FIT | 2SLS |
| Specifies iterated two-stage least squares | FIT | IT2SLS |
| Specifies three-stage least squares | FIT | 3SLS |
| Specifies iterated three-stage least squares | FIT | IT3SLS |
| Specifies full information maximum likelihood | FIT | FIML |
| Specifies simulated method of moments | FIT | NDRAW |
| Specifies number of draws for the V matrix | FIT | NDRAWV |
| Specifies number of initial observations for SMM | FIT | NPREOBS |
| Selects the variance-covariance estimator used for FIML | FIT | COVBEST= |
| Specifies generalized method of moments | FIT | GMM |
| Specifies the kernel for GMM and ITGMM | FIT | KERNEL= |
| Specifies iterated generalized method of moments | FIT | ITGMM |
| Specifies the type of generalized inverse used for the covariance matrix | FIT | GINV= |
| Specifies the denominator for computing variances and covariances | FIT | VARDEF= |
| Specifies adding the variance adjustment for SMM | FIT | ADJSMMV |
| Specifies variance correction for heteroscedasticity | FIT | HCCME= |
| Specifies GMM variance under arbitrary weighting matrix | FIT | GENGMMV |
| Specifies GMM variance under optimal weighting matrix | FIT | NOGENGMMV |
| Solution Mode Options | ||
| Selects a subset of the model equations | SOLVE | SATISFY= |
| Solves only for missing variables | SOLVE | FORECAST |
| Solves for all solution variables | SOLVE | SIMULATE |
| Solution Mode Options: Lag Processing | ||
| Uses solved values in the lag functions | SOLVE | DYNAMIC |
| Uses actual values in the lag functions | SOLVE | STATIC |
| Produces successive forecasts to a fixed forecast horizon | SOLVE | NAHEAD= |
| Selects the observation to start dynamic solutions | SOLVE | START= |
| Solution Mode Options: Numerical Methods | ||
| Specifies the maximum number of iterations allowed | SOLVE | MAXITER= |
| Specifies the maximum number of subiterations allowed | SOLVE | MAXSUBITER= |
| Specifies the convergence criteria | SOLVE | CONVERGE= |
| Computes a simultaneous solution using a Jacobi-like iteration | SOLVE | JACOBI |
| Computes a simultaneous solution using a Gauss-Seidel-like iteration | SOLVE | SEIDEL |
| Computes a simultaneous solution using Newton’s method | SOLVE | NEWTON |
| Computes a nonsimultaneous solution | SOLVE | SINGLE |
| Monte Carlo Simulation Options | ||
| Specifies quasi-random number generator | SOLVE | QUASI= |
| Specifies pseudo-random number generator | SOLVE | PSUEDO= |
| Repeats the solution multiple times | SOLVE | RANDOM= |
| Initializes the pseudo-random number generator | SOLVE | SEED= |
| Specifies copula options | SOLVE | COPULA= |
| Solution Mode Printing Options | ||
| Prints between data points integration values for the DERT. variables and the auxiliary variables | FIT, SOLVE, MODEL | INTGPRINT |
| Prints the solution approximation and equation errors | SOLVE | ITPRINT |
| Prints the solution values and residuals at each observation | SOLVE | SOLVEPRINT |
| Prints various summary statistics | SOLVE | STATS |
| Prints tables of Theil inequality coefficients | SOLVE | THEIL |
| Specifies all printing control options | SOLVE | PRINTALL |
| General TEST Statement Options | ||
| Specifies that a Wald test be computed | TEST | WALD |
| Specifies that a Lagrange multiplier test be computed | TEST | LM |
| Specifies that a likelihood ratio test be computed | TEST | LR |
| Request all three types of tests | TEST | ALL |
| Specifies the name of an output SAS data set that contains the test results | TEST | OUT= |
| Miscellaneous Statements | ||
| Specifies the range of observations to be used | RANGE | |
| Subsets the data set with BY variables | BY | |
Last updated: December 20, 2021