Processes | Expression Parameters | PROC GLIMMIX Options

 • Displays asymptotic correlation matrix of parameter estimates. It is computed from the corresponding asymptotic covariance matrix (see the description of the ASYCOV option, below)
 • This option requests that the asymptotic covariance matrix of the covariance parameters be displayed. By default, this matrix is the observed inverse Fisher information matrix, which equals 2 H -1, where H is the Hessian (second derivative) matrix of the objective function.
 • Computes the estimated -covariance matrix of the fixed-effects parameters by using the asymptotically consistent (or sandwich) estimator.
 • The GLIMMIX procedure normally computes various IC that typically apply a penalty to the (possibly restricted) log likelihood, log pseudo-likelihood, or log quasi-likelihood that depends on the number of parameters and/or the .
 •
 • The GLIMMIX procedure normally computes various IC that typically apply a penalty to the (possibly restricted) log likelihood, log pseudo-likelihood, or log quasi-likelihood that depends on the number of parameters and/or the sample size.
 • Select this option to request that the penalties include the number of fixed-effects parameters, when estimation in models with is based on a residual (restricted) likelihood.
: For METHOD=MSPL , METHOD=MMPL , METHOD=LAPLACE , and METHOD=QUAD , the IC=Q and IC=PQ options produce the same results.
 • The GLIMMIX procedure normally computes various IC that typically apply a penalty to the (possibly restricted) log likelihood, log pseudo-likelihood, or log quasi-likelihood that depends on the number of parameters and/or the sample size.
 • This is the default option for linear with normal errors, and the resulting information criteria are identical to the IC option specified using .
: For METHOD=MSPL , METHOD=MMPL , METHOD=LAPLACE , and METHOD=QUAD , the IC=Q and IC=PQ options produce the same results.
 •
 • Displays the parameter values at each iteration and enables the writing of notes to the SAS log pertaining to infinite likelihood and singularities during Newton-Raphson iterations.
 •
 • : This option was designed for use with analyses requiring extensive CPU resources.
 •
 • The RSPL option specifies that the estimation is based on a R esidual likelihood with a S ubject-specific expansion locus. The PL abbreviation identifies the method as a pseudo-likelihood technique.
 •
 •
 • The MSPL option specifies that the estimation is based on a M aximum likelihood ( R ) with a S ubject-specific expansion locus. The PL abbreviation identifies the method as a pseudo-likelihood technique.
 •
 • The RMPL option specifies that the estimation is based on a R esidual likelihood with a M arginal-specific expansion locus. The PL abbreviation identifies the method as a pseudo-likelihood technique.
 •
 • The MMPL option specifies that the estimation is based on a M aximum likelihood with a M arginal-specific expansion locus. The PL abbreviation identifies the method as a pseudo-likelihood technique.
 •
 • Twice the negative of the resulting log -likelihood approximation is the objective function that the procedure minimizes to determine parameter estimates. Laplace estimates typically exhibit better asymptotic behavior and less small-sample bias than pseudo-likelihood estimators. On the other hand, the class of models for which a Laplace approximation of the marginal log likelihood is available is much smaller compared to the class of models to which PL estimation can be applied.
 • Compared to METHOD=LAPLACE , the models for which parameters can be estimated by quadrature are further restricted.
 •
 • For example, variance components have a default lower boundary constraint of 0, and the NOBOUND option allows their estimates to be negative.
 •
 •
 •
 •
 •
 • Requests that the starting values for the not be obtained by first fitting a generalized linear model.
 •
 •
 •
 •
 •
 •
 • Specifies that the levels of the classification are sorted in the order in which they appear in the input data set.
 •
 •
 •
 •
 •
: In generalized linear models with normally distributed data, you can use the PROFILE option to request profiling of the residual variance.
 •
 To specify more than one option, hold down Ctrl as you left-click on the desired options.