Regression Variables

Most of the Estimation Instructions define a common set of variables. The following are defined for almost any estimation:

%BETA	coefficient VECTOR
%BETASYS	coefficient vector: for certain types of estimation, this includes terms omitted from %BETA (VECTOR)
%LOGL	Normal (Gaussian) log likelihood where appropriate (REAL)
%NFREE	number of free parameters including variances (INTEGER)
%NOBS	number of observations (INTEGER)
%NREG	number of regressors (INTEGER)
%STDERRS	VECTOR of coefficient standard errors
%TSTATS	VECTOR containing the t-stats for the coefficients
%XX	inverse(\({\bf{X'X}}\)) matrix (OLS regressions) or estimated covariance matrix (other estimations) (SYMMETRIC)
%XXSYS	inverse(\({\bf{X'X}}\)) or estimated covariance matrix: For certain types of estimation, this includes terms omitted from %XX (SYMMETRIC)

In addition, the following are defined for most estimation methods with a single dependent variable.

%MEAN	mean of dependent variable (REAL)
%NDF	number of degrees of freedom (INTEGER)
%RESIDS	SERIES containing the residuals
%VARIANCE	variance of dependent variable (REAL)

and the following are also defined for estimations that are univariate and minimize the sum of squared residuals

%DURBIN	Durbin-Watson statistic (REAL)
%RBARSQ	R-bar-squared (REAL)
%RHO	first lag correlation coefficient (REAL)
%RSQUARED	R-squared (REAL)
%RSS	residual sum of squares (REAL)
%SEESQ	standard error of estimate squared (REAL)
%SIGMASQ	maximum likelihood estimate of residual variance (REAL)
%TRSQ	number of observations times raw R-squared (REAL)
%TRSQUARED	number of observations times the centered R-squared (REAL)