INFORMATION SHAZAM? Features Requirements Download Trial Prices UPDATES Account Updates ORDER SHAZAM Buy Online Partners / Resellers Order by Email or Fax Student Discounts Sublicensing ORDER MANUALS The SHAZAM Store Lulu.com Or by Email or Fax RESOURCES Reference Manual Command Reference Examples - Manual Examples - Books FAQs Whistler's Introduction SHAZAM on a Mac SHAZAM at REPEC Manuals & Handbooks

AUTO

The AUTO command provides numerous features for estimation of models with autocorrelated errors or moving average errors. The order of the autoregressive process or the moving average process is specified with the ORDER= option on the AUTO command.

In general, the format is:

AUTO depvar indeps / options

The available options are:

 OPTION DESCRIPTION anova Prints the ANalysis Of VAriance tables and the F-statistic for the test that all coefficients are zero. In some restricted models, the F-test from an ANOVA table is invalid. In these cases the F-statistic will not be printed, but may be obtained with the TEST command. Model Selection Tests are also printed. beg= Specifies the BEGinning observation to be used in estimation. This option overrides the SAMPLE command and defaults to the sample range in effect. coef= Saves the COEFficients in the variable specified. If there is an intercept it will be stored as the last coefficient. conv= Used to set a CONVergence criterion. The default is .001. This option is ineffective for a grid search. cov= Saves the COVariance matrix of coefficients in the variable specified. dlag Used with first-order models to tell SHAZAM that the first independent variable is a LAGged Dependent variable. With this option the estimated variances will be calculated using Dhrymes [1971, Theorem 7.1] dn Computes the estimated variance of the regression line by Dividing the residual sum of squares by N instead of N-K. If the user believes that the model only has good large-sample properties and wants the maximum likelihood estimates of the variances, this option should be used. drop DROPs the first observation in estimation. (In a second-order model, the first two observations will be dropped.) While this yields less efficient estimates, there are cases where this option might be chosen. If this option is not used, the beginning observations are saved with the usual transformation. dump DUMPS large amounts of output mainly useful to SHAZAM Consultants end= Specifies the ENDing observation to be used in estimation. This option overrides the SAMPLE command and defaults to the sample range in effect. gap= Provides an alternative way to indicate a single GAP in the data. It is used when the missing observations were not accounted for in the data. You should specify the observation number immediately before the gap. For example, GAP=29 means that a gap exists in the data after observation 29. It will be assumed that only one observation is missing, unless the number of missing observations is specified with the NMISS= 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. gs Uses a Grid Search to estimate rho. For models with AR(1) errors 39 iterations will be required by a grid search. For models with AR(2) errors over 100 iterations will be required. The grid search will yield an accuracy of 0.01. iter= May be used when doing an ITERative Cochrane-Orcutt procedure to control the number of iterations. If a value of ITER= is specified, this will be the maximum number of iterations allowed. ITER= should be a number between 2 and 99. The default is 19. lininv Used when the dependent variable is LINear, but the independent variables are in INVerse form. This type of model is known as the reciprocal model. This option only affects the calculation of the elasticities. NOTE: This option does not transform the data. The data must be transformed by the user with the appropriate GENR commands or the Data Editor. See the chapter ORDINARY LEAST SQUARES in the SHAZAM Reference Manual. linlog Used when the dependent variable is LINear, but the independent variables are in LOG form. This option only affects the calculation of the elasticities. NOTE: This option does not transform the data. The data must be transformed by the user with the appropriate GENR commands or the Data Editor. See the chapter ORDINARY LEAST SQUARES in the SHAZAM Reference Manual. 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. loginv Used when the dependent variable is in LOG form, and the independent variables are in INVerse form. loglin Used when the dependent variable is in LOG form, but the independent variables are LINear. This option only affects the calculation of the elasticities. NOTE: This option does not transform the data. The data must be transformed by the user with the appropriate GENR commands or the Data Editor. See the chapter ORDINARY LEAST SQUARES in the SHAZAM Reference Manual. loglog Used when the dependent variable is in LOG form, and the independent variables are in LOG form. This option only affects the calculation of the elasticities. NOTE: This option does not transform the data. The data must be transformed by the user with the appropriate GENR commands or the Data Editor. See the chapter ORDINARY LEAST SQUARES in the SHAZAM Reference Manual. 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. miss Adjusts the maximum likelihood estimates for any MISSing observations that were deleted with SKIPIF commands. ml Does Maximum Likelihood estimation for models with first or second order autoregressive errors. For the model with AR(1) errors the estimation method is a modified Cochrane-Orcutt procedure as developed by Beach and MacKinnon [1978]. With the GS option a grid search is used. A model with AR(2) errors is specified with the ORDER=2 option and the estimation uses a grid search. Note that, with the ML option, the DN option will automatically be in effect and the DROP option may not be used. nmiss= Used with the GAP= option to indicate the Number of MISSing observations from the data. For example, GAP=29 NMISS=3 means that 3 observations were missing from the data after observation 29. This option is not to be confused with the MISS option described above. noanova Specifies not to print Analysis of Variance tables and model selection tests noconstant There will be NO CONSTANT (intercept) in the estimated equation. This option is used when the intercept is to be suppressed in the regression or when the user is supplying the intercept. This option should be used with caution as some of the usual output may be invalid. In particular, the usual R^2 is not well defined and could be negative. However, when this option is used, the raw moment R^2 may be of interest. The ANALYSIS OF VARIANCE - FROM MEAN table will not be computed if this option is used. nopiter Suppresses intermediate output from iterations. nowide Uses a narrower page width for the display of results. For printing, the page set-up can be set to portrait orientation. numarma= Specifies the NUMber of AR or MA coefficients to be estimated if some of the coefficients are to be restricted to be zero. For example, when ORDER=4 is specified, but it is desired that the second autoregressive coefficient should be zero and not estimated then use NUMARMA=3 and in the line following the AUTO command include the 3 coefficents to be estimated. If the NUMARMA= option is used with ORDER=2 then the PAGAN option must also be specified. order= Used to estimate models with second or higher-ORDER autoregressive errors. If ORDER=2 is specified, an iterative Cochrane-Orcutt procedure is done unless the GS or ML options are also specified. The missing data options are not valid for ORDER=1 estimation. If an ORDER larger than 2 is specified the model will be estimated using the least squares procedure described in Pagan [1974]. The DLAG, DROP, GAP, GS, MISS, ML, NMISS, RESTRICT, RHO and SRHO options are not permitted when ORDER is larger than 2. Also, the FC command will not work after an AUTO command that uses this option. If a negative ORDER is specified the model will be estimated using a moving-average error model instead of an autoregressive error model. A least squares procedure is used as described in Pagan [1974]. For example, to estimate a model with first-order moving average errors use the option ORDER=-1. The MISS, GAP, RHO, SRHO, DLAG, DROP, GS, ML and RESTRICT options are not permitted. The FC command will not work after an AUTO command that uses this option. pagan May be used with ORDER=1 or ORDER=2 to estimate the model using Pagan's [1974] procedure. pcor Prints the CORrelation matrix of the estimated coefficients. This should not be confused with a correlation matrix of variables which can be obtained with a STAT command. pcov Prints the COVariance matrix of coefficients. This should not be confused with the covariance matrix of variables which can be obtained with the STAT command. predict= Saves the PREDICTed values of the dependent variable in the variable specified. resid= Saves the values of the RESIDuals from the regression in the variable specified. restrict Forces linear RESTRICTions into the regression. It tells SHAZAM that RESTRICT commands follow. Restrictions must be linear. This option may not be used with DWPVALUE or METHOD=HH. rho= Allows the specification of any value of rho desired for the regression. With the RHO= option neither a Cochrane-Orcutt nor a maximum likelihood estimation is done since SHAZAM already has the value of rho. This is useful when rho is already known as it eliminates expensive iterations. 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. srho= Used with the ORDER=2 and RHO= options when the Second-order p is to be specified for a model with AR(2) errors. stderr= 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. wide Uses a wider page width for the display of results. For printing, the page set-up should be set to landscape orientation.

The available temporary variables on the AUTO command are:

\$ADR2, \$DF, \$ERR, \$K, \$LLF, \$N, \$R2, \$RAW, \$SIG2, \$SSE, \$SSR, \$SST, \$ZDF, \$ZSSR, and \$ZSST

With the RSTAT option the available temporary variables are:

\$DURH, \$DW, \$R2OP and \$RHO