| OPTION |
DESCRIPTION |
| coef= |
Saves the values of the COEFficients in the variable specified. The values for all equations will be in a single vector. No values for the equation intercepts will be saved so if these are required the directions above for the NOCONSTANT option should be followed. |
| coefmat= |
Saves the values of the COEFficients in the variable specified in MATrix form with each column of the matrix representing one equation. Equation intercepts will also be saved. This is useful if the coefficients for a single equation need to be specified on the FC command. |
| conv= |
Specifies a CONVergence criterion to stop the iterative procedure (subject to the maximum number of iterations specified with the ITER= option). The criterion is the maximum desired percentage change in each of the coefficients. The default is 0.001. |
| cov= |
Saves the COVariance matrix in the variable specified. No values for the variances and covariances of equation intercepts will be saved so if these are required the directions above for the NOCONSTANT option should be followed. |
| dn |
The covariance matrix is computed using N as the Divisor and is strongly recommended. Without the DN option, the degrees of freedom correction for the covariances is as described in the SHAZAM Reference Manual. Some authors prefer this Divisor over other Degrees of Freedom adjustments. |
| dump |
DUMPs a lot of intermediate output after each iteration, including the system X'X matrix and its inverse, and the inverse of the residual covariance matrix (SIGMA). This option is primarily of interest to SHAZAM consultants. |
| full |
Lists FULL equation output. This output is similar to that of regular OLS regressions. FULL is automatically in effect except in TALK mode. It can be turned off with the NOFULL option. |
| gf |
Prints Goodness of Fit tests for normality of residuals. Coefficients of skewness and excess kurtosis and the Jarque-Bera test for normality of the residuals are also computed. Users should be aware when interpreting the equation-by-equation output that some statistics printed are not valid in system estimation. If systems analysis is properly understood the questionable statistics are easily identified. |
| in= |
This option is used with the OUT= option. These options are used to OUTput a dump of useful information on the unit specified at each iteration, so that the system can be restarted in another run at the same point by INputting the dump. This can be very useful in expensive models to avoid re-estimation of already calculated data in the event that a time limit is reached. OUT= and IN= are usually assigned to the same unit so the latest information replaces existing information. Units 11-49 are available and may be assigned to a file with the SHAZAM FILE command or an operating system command. It is also important to run the same system on subsequent runs with the IN= option. The IN= option should never be run before there is anything to input. |
| iter= |
Specifies the maximum number of ITERations performed if an iterative procedure is desired. If this option is not specified, one iteration is done. If ITER=0 is specified, the system is estimated without the Generalized Least Squares procedure which uses the covariance matrix of residuals. This would be equivalent to running separate OLS regressions (or 2SLS), but more expensive. ITER=0 would be appropriate only if there were restrictions across equations or hypothesis testing across equations. |
| list |
LISTs and plots the residuals and predicted values of the dependent variable and residual statistics. When LIST is specified RSTAT is automatically turned on. Users should be aware when interpreting the equation-by-equation output that some statistics printed are not valid in system estimation. If systems analysis is properly understood the questionable statistics are easily identified. |
| max |
Prints Analysis of Variance Tables, Variance-covariance matrix, Correlation matrix, Residuals, Residual Statistics and Goodness of Fit Test for Normality. This option is equivalent to using the ANOVA, LIST, PCOV, PCOR and GF options. Users should be sure the MAX output is necessary, otherwise unnecessary calculations are required. Users should be aware when interpreting the equation-by-equation output that some statistics printed are not valid in system estimation. If systems analysis is properly understood the questionable statistics are easily identified. |
| noconexog |
If a list of exogenous variables is included for either Two or Three Stage Least Squares SHAZAM will automatically add a CONSTANT to the list. If you do not want SHAZAM to automatically include a constant in the list of exogenous variables, specify the NOCONEXOG option. |
| noconstant |
Normally, SHAZAM will automatically put an intercept in each equation in the system. If there are some (or all) equations in which no intercept is desired the NOCONSTANT option should be specified. Then, an intercept should be created by generating a variable of ones (1) with a GENR command and this variable should be included in each equation in which an intercept is desired. Without the NOCONSTANT option the intercept will only be printed in the FULL equation-by-equation output, and the variances of the intercepts will only be approximate unless the model converges. |
| nofull |
Disables FULL equation output. This output is similar to that of regular OLS regressions. FULL is automatically in effect except in TALK mode. It can be turned off with the NOFULL option |
| nogf |
Specifies not to print Goodness of Fit tests for normality of residuals, skewness, kurtosis and the Jarque-Bera test |
| out= |
This option is used with the IN= option. These options are used to OUTput a dump of useful information on the unit specified at each iteration, so that the system can be restarted in another run at the same point by INputting the dump. This can be very useful in expensive models to avoid re-estimation of already calculated data in the event that a time limit is reached. With the OUT= option, the information from the most recent iteration will be written on the specified unit. OUT= and IN= are usually assigned to the same unit so the latest information replaces existing information. Units 11-49 are available and may be assigned to a file with the SHAZAM FILE command or an operating system command. It is important to remember to create a system file to be attached to the appropriate unit before using the OUT= option. Without this file the information will be lost. |
| pcor |
Prints the CORrelation matrix of all coefficients in the system after convergence. If these options are specified the matrices will also be printed for each equation in the equation-by-equation output. |
| pcov |
Prints the COVariance matrix of all coefficients in the system after convergence. If these options are specified the matrices will also be printed for each equation in the equation-by-equation output. |
| pinvev |
Prints the INVerse of the Exogenous Variables matrix (X`X)^-1 when 3SLS estimation is being used. This option is rarely needed. |
| piter= |
Specifies the frequency with which ITERations are to be Printed. The default is PITER=1. If PITER=0 is specified, no iterations are printed. |
| predict= |
Saves the PREDICTed values of the dependent variable in a N x M matrix, where N is the number of observations and M is the number of equations. |
| psigma |
Prints the residual covariance matrix (SIGMA) after each iteration. If this option is not used the matrix will be printed for the first and final iterations. This option can add pages of output if there are many equations in the system and, thus, should not be specified unless this information is specifically needed. |
| resid= |
Saves the values of the estimated RESIDuals in a N x M matrix. |
| restrict |
Forces linear restrictions as specified with RESTRICT commands. The RESTRICT commands follow the OLS commands. Only linear restrictions are permitted. Users should be aware that iterative estimates may not necessarily be maximum likelihood estimates if restrictions on the intercepts of the equation are imposed. |
| rstat |
Prints Residual Summary STATistics. The output includes the Durbin-Watson statistic and related residual test statistics. It also includes the Runs Test. When the LIST option is specified RSTAT is automatically turned on. Users should be aware when interpreting the equation-by-equation output that some statistics printed are not valid in system estimation. If systems analysis is properly understood the questionable statistics are easily identified. |
| sigma= |
Save the residual covariance matrix (SIGMA) in the variable specified. |
| stderror= |
Saves the values of the STanDard ERRors of the coefficients in the variable specified. |
| tratio= |
Saves the values of the T-RATIOs in the variable specified. |
